{"id":"10087359-956f-4ade-8d48-adce21ae1b09","entityType":"agent","slug":"clawhub-deciqai-deliberate-practice","name":"Deliberate Practice","canonicalUrl":"https://www.xpersona.co/agent/clawhub-deciqai-deliberate-practice","canonicalPath":"/agent/clawhub-deciqai-deliberate-practice","generatedAt":"2026-10-11T08:41:39.360Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-11T06:20:09.997Z","emptyReason":null},"description":"Activate when: user says 'I've been doing this for years but I'm not getting better'; someone suspects a skill plateau despite continued effort; designing a... Skill: Deliberate Practice Owner: deciqai Summary: Activate when: user says 'I've been doing this for years but I'm not getting better'; someone suspects a skill plateau despite continued effort; designing a... Tags: latest:1.0.6 Version history: v1.0.6 | 2026-07-16T17:57:13.883Z | user Description tail link + agents machine-readable metadata line (deciqai.com/s/deliberate-practice.json) v1.0.5 | 2026-07-10T10:24:58.","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 1.1K downloads reported by the source. Last updated 10/11/2026.","installCommand":"clawhub skill install s17a4mqcnk515kvaca5ze55d0x88pfpx:deliberate-practice","sourceUrl":"https://clawhub.ai/deciqai/deliberate-practice","homepage":"https://clawhub.ai/deciqai/skills/deliberate-practice","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/deciqai/deliberate-practice","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/deciqai/skills/deliberate-practice","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":61,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"Activate when: user says 'I've been doing this for years but I'm not getting better'; someone suspects a skill plateau despite continued effort; designing a... "},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-11T06:20:09.997Z","emptyReason":null},"protocols":[{"protocol":"OPENCLEW","label":"OpenClaw","status":"self-declared","notes":"Declared in the public agent profile."}],"capabilities":[],"verifiedCount":0,"selfDeclaredCount":1,"capabilityMatrix":{"rows":[{"key":"OPENCLEW","type":"protocol","support":"unknown","confidenceSource":"profile","notes":"Listed on profile"}],"flattenedTokens":"protocol:OPENCLEW|unknown|profile"}},"adoption":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T06:20:09.997Z","emptyReason":null},"stars":null,"forks":null,"downloads":1135,"packageName":null,"latestVersion":"1.0.6","tractionLabel":"1.1K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T06:20:09.984Z","emptyReason":null},"lastUpdatedAt":"2026-10-11T06:20:09.997Z","lastCrawledAt":"2026-10-11T06:20:09.984Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-12T06:20:09.984Z","lastVerifiedAt":null,"highlights":[{"version":"1.0.6","createdAt":"2026-07-16T17:57:13.883Z","changelog":"Description tail link + agents machine-readable metadata line (deciqai.com/s/deliberate-practice.json)","fileCount":7,"zipByteSize":14627},{"version":"1.0.5","createdAt":"2026-07-10T10:24:58.498Z","changelog":"Add 2024-2026 AI-era worked example + updated sources","fileCount":7,"zipByteSize":14473},{"version":"1.0.4","createdAt":"2026-07-08T10:59:20.733Z","changelog":"Footer now uses /c/<slug> short link (fixes UTM truncation when SKILL.md is read in a terminal)","fileCount":6,"zipByteSize":10252},{"version":"1.0.3","createdAt":"2026-07-08T03:51:31.680Z","changelog":"Second primary-sourced worked example","fileCount":6,"zipByteSize":10313},{"version":"1.0.2","createdAt":"2026-07-08T00:44:15.773Z","changelog":"Refreshed content + GitHub star link in footer","fileCount":5,"zipByteSize":7771},{"version":"1.0.1","createdAt":"2026-07-07T20:31:58.245Z","changelog":"Add catalog categories and topics","fileCount":5,"zipByteSize":7589},{"version":"1.0.0","createdAt":"2026-06-27T07:17:51.993Z","changelog":"Initial publish","fileCount":5,"zipByteSize":7417}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s17a4mqcnk515kvaca5ze55d0x88pfpx:deliberate-practice","setupComplexity":"low","setupSteps":["Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.","Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data."],"contract":{"contractStatus":"missing","authModes":[],"requires":[],"forbidden":[],"supportsMcp":false,"supportsA2a":false,"supportsStreaming":false,"inputSchemaRef":null,"outputSchemaRef":null,"dataRegion":null,"contractUpdatedAt":null,"sourceUpdatedAt":null,"freshnessSeconds":null},"invocationGuide":{"preferredApi":{"snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-deliberate-practice/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-deliberate-practice/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-deliberate-practice/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-deliberate-practice/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-deliberate-practice/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-deliberate-practice/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-11T08:41:39.353Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-deliberate-practice/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-deliberate-practice/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-deliberate-practice/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-deliberate-practice/trust"}},"reliability":{"evidence":{"source":"runtime-metrics","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No trust, reliability, or runtime telemetry is available."},"trust":{"status":"unavailable","handshakeStatus":"UNKNOWN","verificationFreshnessHours":null,"reputationScore":null,"p95LatencyMs":null,"successRate30d":null,"fallbackRate":null,"attempts30d":null,"trustUpdatedAt":null,"trustConfidence":"unknown","sourceUpdatedAt":null,"freshnessSeconds":null},"decisionGuardrails":{"doNotUseIf":["Contract metadata is missing or unavailable for deterministic execution."],"safeUseWhen":[],"riskFlags":["missing_or_unavailable_contract","trust_data_unavailable","schema_references_missing"],"operationalConfidence":"low"},"executionMetrics":{"observedLatencyMsP50":null,"observedLatencyMsP95":null,"estimatedCostUsd":null,"uptime30d":null,"rateLimitRpm":null,"rateLimitBurst":null,"lastVerifiedAt":null,"verificationSource":null},"runtimeMetrics":{"successRate":null,"avgLatencyMs":null,"avgCostUsd":null,"hallucinationRate":null,"retryRate":null,"disputeRate":null,"p50Latency":null,"p95Latency":null,"lastUpdated":null}},"benchmarks":{"evidence":{"source":"no-benchmark-data","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No benchmark suites or observed failure patterns are available."},"suites":[],"failurePatterns":[]},"artifacts":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"high","updatedAt":"2026-10-11T06:20:09.997Z","emptyReason":null},"readme":"Skill: Deliberate Practice\n\nOwner: deciqai\n\nSummary: Activate when: user says 'I've been doing this for years but I'm not getting better'; someone suspects a skill plateau despite continued effort; designing a...\n\nTags: latest:1.0.6\n\nVersion history:\n\nv1.0.6 | 2026-07-16T17:57:13.883Z | user\n\nDescription tail link + agents machine-readable metadata line (deciqai.com/s/deliberate-practice.json)\n\nv1.0.5 | 2026-07-10T10:24:58.498Z | user\n\nAdd 2024-2026 AI-era worked example + updated sources\n\nv1.0.4 | 2026-07-08T10:59:20.733Z | user\n\nFooter now uses /c/<slug> short link (fixes UTM truncation when SKILL.md is read in a terminal)\n\nv1.0.3 | 2026-07-08T03:51:31.680Z | user\n\nSecond primary-sourced worked example\n\nv1.0.2 | 2026-07-08T00:44:15.773Z | user\n\nRefreshed content + GitHub star link in footer\n\nv1.0.1 | 2026-07-07T20:31:58.245Z | user\n\nAdd catalog categories and topics\n\nv1.0.0 | 2026-06-27T07:17:51.993Z | user\n\nInitial publish\n\nArchive index:\n\nArchive v1.0.6: 7 files, 14627 bytes\n\nFiles: examples/berlin-violin-study-1991-1993.md (2172b), examples/franklin-spectator-writing-method.md (4709b), examples/skill-atrophy-when-ai-does-the-reps-2024-2026.md (6741b), references/sources.md (3019b), skill-card.md (2890b), SKILL.md (8289b), _meta.json (138b)\n\nFile v1.0.6:SKILL.md\n\n---\nname: deliberate-practice\ndescription: \"Activate when: user says 'I've been doing this for years but I'm not getting better'; someone suspects a skill plateau despite continued effort; designing a learning program for a high-performance outcome; an organization reports high training hours but low skill transfer.\n  Do NOT activate when: goal is execution of existing skills rather than acquiring new ones (use deep-work instead); there is no identifiable expert performance benchmark to target. More: deciqai.com/c/deliberate-practice\"\n---\n\n# Deliberate Practice\n\n## Overview\n\nMost people confuse repetition with learning — they accumulate years of experience and plateau. Once an activity becomes automatic, executing it no longer builds new neural architecture. Ericsson, Krampe & Tesch-Römer (1993) showed the predictive variable is not hours of doing but hours of *specifically deliberate practice* — targeted, uncomfortable, feedback-rich repetition designed to build mental representations.\n\n**Cross-skill composition:** Use `feedback-loops` first (audit your error signal); then `metacognition` (surface your current representation gap); use instead of `deep-work` when acquiring skills, not producing output; use alongside `cognitive-evolution-stages` for stage-aware practice design.\n\n---\n\n## When to Use\n\n**Trigger:** plateau despite experience; designing high-performance learning program; training hours high but skill transfer low; evaluating whether practice is building capability or maintaining it; skill atrophy or deskilling as AI copilots absorb the routine reps (AI adoption, AI hype, \"will AI make me worse at my craft\").\n**When NOT:** goal is execution not acquisition (use deep-work); no expert benchmark exists; bottleneck is motivational not representational.\n\n---\n\n## Coaching Novices (Adaptive Front Door)\n\n**Engine mode:** user has a concrete case → run The Process directly.\n**Coach mode:** user is unfamiliar or has no concrete case → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. Ask the plateau question: \"When you practice X, what does it feel like after 15 minutes — harder, the same, or easier?\" Comfort/easy = automatic = not building representations.\n2. Find the expert performance structure: \"Who is world-class at X? What do they perceive in the first 3 seconds that you don't?\" This locates the mental representation gap.\n3. Identify the discomfort zone: \"What part of practicing X makes you most want to stop?\" That is almost always where the gap lives.\n> **[WAIT — do not advance until user responds]**\n4. Design the smallest feedback loop: \"How would you know within 60 seconds whether a move was correct?\" Latency over 24h kills representation-building.\n> **[WAIT — do not advance until user responds]**\n5. Set the repetition target and stop-rule: \"How many reps of this specific discomfort can you sustain before concentration drops?\" (1–4 hours/day is Ericsson's ceiling.)\n> **[WAIT — do not advance until user responds]**\n\n---\n\n## The Process\n\n**Step 1 — Define the sub-skill with precision.** Not \"get better at X\" — specify the exact representational gap (e.g., \"detect when counterpart shifts from positional to interest-based\").\n**Step 2 — Find or construct the feedback mechanism.** Latency >24h breaks action-result association. Expert feedback > peer one level above > simulation with ground truth.\n**Step 3 — Diagnose the mental representation gap.** Ask: \"What does an expert *see* here that I don't?\" Not what they do — the doing follows from the seeing.\n**Step 4 — Design the repetition targeting the gap.** Must trigger the sub-skill, produce in-session feedback, and be executable at dozens–hundreds of reps per session.\n**Step 5 — Track representation progress, not output.** Output metrics lag by weeks. Track: \"Am I perceiving X earlier than before?\"\n**Step 6 — Apply the stop-rule.** Comfort = automaticity maintenance. Redesign to a harder sub-skill. End session when concentration drops.\n\n### Output: Practice Design Artifact\n\n```\nTarget sub-skill (precise): [specific representational gap]\nExpert mental representation: [what expert perceives that I currently don't]\nCurrent representation gap: [specific failure mode]\nFeedback — Source / Latency / Reliability:\nRepetition — Exercise / Volume / Duration / Frequency:\nProgress indicator (representation-level, not output): [what I will perceive by Week N]\nStop-rule triggers: comfortable → redesign; concentration drops → end; latency >24h → redesign\n```\n\n*→ Method in Action: [Berlin Violin Study (1991–1993)](examples/berlin-violin-study-1991-1993.md) · [Franklin's Spectator Method](examples/franklin-spectator-writing-method.md)*\n*→ 2026 lens: [Keeping Skill Alive When AI Does the Reps (2024–2026)](examples/skill-atrophy-when-ai-does-the-reps-2024-2026.md)*\n\n---\n\n## Practice Design Domain Packs\n\n**Medicine/Surgery:** sub-skill: laparoscopic tissue manipulation; feedback: simulator + debrief within 1h; rationalization to reject: \"I'll improve with more cases.\"\n**Writing:** sub-skill: eliminate nominalization in first-draft prose; feedback: rewrite published paragraphs vs original; rationalization: \"I write every day.\"\n**Investment:** sub-skill: identify customer concentration risk from footnotes in 20 min; feedback: 50-case retrospective library with outcomes.\n\nContribute packs via the deciqAI repo — requires sub-skill, expert representation, feedback latency, and common rationalization.\n\n---\n\n## Applying It Well\n\n- Target representations, not outcomes — ask \"What does the expert *perceive* that I don't?\"\n- Make feedback faster — redesign question: \"How do I get a reliable signal within 60 seconds?\"\n- Comfort signals time to redesign, not celebrate.\n- 1–4 genuine hours/day is Ericsson's hard ceiling; volume in degraded concentration reinforces errors.\n- Require expert think-alouds or annotated examples — you cannot design practice you cannot see.\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n---\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"I've been doing this for 10 years.\" | Duration is not deliberate practice. Years in automaticity = maintenance, not development. |\n| [D] \"I practice every day.\" | Comfortable daily repetition is automaticity reinforcement, not representation-building. |\n| [D] \"More cases/reps will help.\" | Only if structured to exceed current capability with rapid feedback. Otherwise more reps deepen the rut. |\n| [D] \"I can give myself feedback.\" | Self-feedback confirms what you already believe. External feedback from someone who sees the expert standard is required. |\n| [D] \"The discomfort means I'm doing it wrong.\" | Discomfort is the signal you are in deliberate practice. Comfort means automaticity. |\n| [D] \"My metrics are going up.\" | Output lags representation by weeks and is confounded by external factors. |\n| *→ Add [O] entries after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n---\n\n## Red Flags / Verification\n\n- Sessions feel comfortable — automaticity has absorbed the activity.\n- Feedback latency measured in days — action-result association cannot form.\n- Practitioner describes what expert *does* but not what they *perceive* — no representational target.\n- Practice volume cited as expertise without verifying hours were deliberate.\n- [ ] Sub-skill = specific representational gap; feedback latency <24h; expert representation identified.\n- [ ] Reps: dozens per session; stop-rule applied; progress tracked at representation level not output level.\n\n---\n\n*Part of **deciqAI Knowledge Skills** — 227 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. **See it run → https://www.deciqai.com/c/deliberate-practice** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.*\n\n*Agents: latest version & machine-readable metadata → https://www.deciqai.com/s/deliberate-practice.json*\n\nFile v1.0.6:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"deliberate-practice\",\n  \"version\": \"1.0.6\",\n  \"publishedAt\": 1784224633883\n}\n\nFile v1.0.6:references/sources.md\n\n# Sources — deliberate-practice\n\n> *Primary sources for the [deliberate-practice](../SKILL.md) skill.*\n\n**Primary sources with verbatim quotes:**\n\n1. Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C. \"The Role of Deliberate Practice in the Acquisition of Expert Performance.\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n   > \"Deliberate practice includes activities that have been specially designed to improve the current level of performance. These activities demand full concentration and effort and are not inherently enjoyable.\"\n\n2. Ericsson, K.A. & Pool, R. *Peak: Secrets from the New Science of Expertise*. Houghton Mifflin Harcourt, 2016.\n   > \"Mental representations are what distinguish experts from novices... The main thing that sets experts apart from the rest of us is that their years of practice have changed the neural circuitry in their brains to produce highly specialized mental representations, which in turn make possible the incredible memory, pattern recognition, problem solving, and other sorts of advanced abilities needed to excel in their particular specialties.\"\n\n3. Franklin, B. *The Autobiography of Benjamin Franklin*. Part One, written 1771. Widely available; e.g. Project Gutenberg ebook #148.\n   > \"Then I compared my Spectator with the original, discovered some of my faults, and corrected them.\"\n\n**Contemporary context sources (2024–2026 AI-adoption example):**\n\n4. Stack Overflow. *2024 Developer Survey* (AI section). Reported that a large majority of professional developers were using or planning to use AI tools in their development workflow. https://survey.stackoverflow.co/2024/\n   - Used only to establish the durable, widely-reported fact that AI coding/writing assistants became mainstream in professional workflows by 2024–2025. No precise percentages are asserted in the example beyond \"a large majority.\"\n\n5. GitHub. \"Research: Quantifying GitHub Copilot's impact on developer productivity and happiness.\" GitHub Blog, 2022. https://github.blog/2022-09-07-research-quantifying-github-copilots-impact-on-developer-productivity-and-happiness/\n   - Cited as background evidence for Copilot's productivity effect and its adoption as a routine tool; the deskilling/atrophy risk in the example is a *prediction from Ericsson's model*, not a claim of a measured 2024–2026 effect.\n\n**What is NOT cited and why:**\n\n- Gladwell, M. *Outliers* (2008) is deliberately excluded. Gladwell popularized the 10,000-hour figure while dropping the \"deliberate\" qualifier — the central mechanism of Ericsson's finding. Citing Gladwell would reproduce the distortion this skill is designed to correct.\n- Chase & Simon (1973) \"Perception in Chess\" (*Cognitive Psychology*, 4(1), 55–81) is the foundational precursor study on expert mental representations in chess — relevant but not cited directly because Ericsson's 1993 paper synthesizes and extends this work and is the canonical source for deliberate practice specifically.\n\nFile v1.0.6:examples/berlin-violin-study-1991-1993.md\n\n# Method in Action: Berlin Violin Study (1991–1993)\n\n> *Example for the [deliberate-practice](../SKILL.md) skill.*\n\n**Source:** Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C., \"The Role of Deliberate Practice in the Acquisition of Expert Performance,\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n\nEricsson and colleagues studied violin students at the Musikhochschule (Academy of Music) in West Berlin. Teachers nominated students into three groups based on their assessment of students' potential: (1) the \"best\" violinists — those judged capable of international solo careers; (2) \"good\" violinists — talented but below the top group; (3) music teachers — students training to teach, not perform at elite level.\n\nAll three groups had begun playing at approximately age 5. By age 20:\n\n- **Best violinists** had accumulated an estimated **10,000 hours** of deliberate practice (defined specifically as practice designed to improve performance, not performance itself or music-related activities like theory or group rehearsal).\n- **Good violinists** had accumulated approximately **8,000 hours**.\n- **Music teachers** had accumulated approximately **4,000 hours**.\n\nThe critical methodological point — erased by Malcolm Gladwell's 2008 popularization — is that Ericsson carefully distinguished *deliberate practice* (specifically designed, uncomfortable, feedback-rich, coach-structured) from total music-related activity time. The predictive variable was not \"hours of playing\" but \"hours of specifically deliberate practice.\" The best violinists also rated deliberate practice as significantly less enjoyable than performance, yet they allocated more time to it — evidence that intrinsic enjoyment does not drive deliberate practice; external structure and long-term goal commitment do.\n\nThis study is the primary empirical foundation for understanding expert performance. It shows that the representation gap between good and best performers is not mysterious — it is the cumulative product of structured, targeted, uncomfortable repetition designed to build increasingly sophisticated internal models.\n\nFile v1.0.6:examples/franklin-spectator-writing-method.md\n\n# Method in Action: Benjamin Franklin's Spectator Writing Method (c. 1718–1723)\n\n> *Example for the [deliberate-practice](../SKILL.md) skill.*\n\nAs a teenage printer's apprentice in Boston, Benjamin Franklin engineered — two centuries before Ericsson named the mechanism — a complete deliberate practice system for prose writing, documented in detail in his *Autobiography*. It is the canonical historical case of self-designed practice in a domain with no available coach.\n\n**Step 1 — Define the sub-skill with precision.** Franklin did not resolve to \"write better.\" His father had compared his letters (from a written debate with his friend John Collins) against Collins's and identified specific deficits: Franklin's writing fell short in elegance of expression, in method, and in clarity. Franklin took these as three distinct representational gaps — word choice, arrangement of thoughts, and expression — and attacked each separately.\n\n**Step 2 — Find or construct the feedback mechanism.** With no writing tutor available, Franklin constructed a ground-truth comparator: an odd volume of *The Spectator*, the London periodical of Addison and Steele, whose prose he judged excellent. The original essay itself became the expert benchmark against which every attempt could be scored — feedback latency of minutes, not days.\n\n**Step 3 — Diagnose the mental representation gap.** The exercise structure forced the diagnosis. Franklin made short hints of the sentiment of each sentence in an essay, set them aside for a few days, then attempted to reconstruct the full essay from the hints in his own words. Comparing his reconstruction against Addison's original exposed exactly where his internal model of good prose diverged from the expert's — fault by fault, sentence by sentence.\n\n**Step 4 — Design the repetition targeting the gap.** When comparison revealed his stock of words was too small, Franklin redesigned the drill: he turned Spectator essays into verse and, after forgetting the originals, back into prose — because versification forces a continual search for words of different lengths and sounds to fit meter and rhyme. When the gap was arrangement rather than vocabulary, he jumbled his sentence hints into confusion, waited weeks, and then attempted to reorder them into the best structure before reconstructing — a drill isolating the method-of-thought sub-skill alone.\n\n**Step 5 — Track representation progress, not output.** Franklin's progress measure was perceptual, not productive: he tracked the faults he could now *discover* in his own reconstructions and correct. He records that he sometimes had the pleasure of judging that, in small particulars, he had improved on the original's method or language — evidence his internal representation of good prose had begun to match, and locally exceed, the benchmark.\n\n**Step 6 — Apply the stop-rule.** The escalation from plain reconstruction, to verse conversion, to jumbled-hint reordering is the stop-rule in action: each time an exercise stopped exposing faults, Franklin redesigned toward a harder sub-skill rather than accumulating comfortable reps. He also confined practice to protected windows — before work in the morning and after it at night — rather than diluting it across the day.\n\nThe outcome: Franklin became the most widely read prose stylist in colonial America, and his *Autobiography* passage became a standard exhibit in the modern science of expertise — Ericsson and Pool analyze it in *Peak* (2016) as a model of practice design under the constraint of no teacher.\n\nNote the counter to the \"I write every day\" rationalization: Franklin already spent his working days surrounded by text in the print shop. The day job never closed the gap — the designed, uncomfortable, feedback-rich exercises did.\n\nThe mapped steps:\n1. Sub-skill defined precisely: three named gaps (word stock, arrangement, expression), attacked separately\n2. Feedback mechanism constructed: the Spectator original as ground truth, comparison within the session\n3. Representation gap diagnosed: reconstruct-and-compare exposes divergence fault by fault\n4. Repetition designed to the gap: verse conversion for vocabulary; jumbled hints for structure\n5. Progress tracked at representation level: faults he could newly perceive, not pages produced\n6. Stop-rule applied: exercise redesigned harder whenever it stopped exposing faults\n\nPrimary source: Franklin, B. *The Autobiography of Benjamin Franklin* (Part One, written 1771; widely available, e.g. Project Gutenberg ebook #148). Analyzed in Ericsson, K.A. & Pool, R., *Peak: Secrets from the New Science of Expertise*, Houghton Mifflin Harcourt, 2016.\n\nFile v1.0.6:examples/skill-atrophy-when-ai-does-the-reps-2024-2026.md\n\n# Method in Action: Keeping Skill Alive When AI Does the Reps (2024–2026)\n\n> *Example for the [deliberate-practice](../SKILL.md) skill.*\n\nOver 2024–2025, AI coding assistants, writing copilots, and analysis tools moved from novelty to default, and by 2026 that shift was well underway. GitHub Copilot, ChatGPT, Claude, and similar tools became embedded in daily engineering, writing, and analyst workflows. This created a specific hazard that deliberate-practice theory predicts precisely: when a copilot handles the routine reps, the practitioner stops doing the very repetitions that once — almost accidentally — built and maintained their mental representations. The activity looks the same on the calendar (still \"coding,\" still \"writing\"), but the skill-building substrate has been quietly outsourced.\n\nThe failure mode is not laziness. It is that *offloading the routine also offloads the practice*. A junior analyst who once wrote every SQL query by hand built pattern recognition as a side effect; a copilot that autocompletes the query removes both the tedium and the representation-building. Below is how a deliberate engineer, writer, or analyst applies this skill to keep building expertise even as AI absorbs the routine work.\n\n**Step 1 — Define the sub-skill with precision.** Not \"stay sharp despite AI.\" Name the exact representational gap that the copilot is now filling on your behalf. For an engineer: \"reason about why a proposed implementation is wrong before running it — hold the data flow in my head rather than pasting the error back to the model.\" For a writer: \"structure an argument from a blank page — sequence claims and transitions without a draft to react to.\" For an analyst: \"form a hypothesis about what a dataset will show *before* querying it.\" The gap is always the perception step the AI now performs first.\n\n**Step 2 — Find or construct the feedback mechanism.** The AI itself is a fast ground-truth comparator if used deliberately rather than as a crutch. Attempt the task cold first; then ask the model for its solution and diff it against yours. Latency is seconds — far inside the 24-hour window where action-result association still forms. The critical discipline: generate your own answer *before* seeing the model's, or the comparison degrades into passive reading and the representation never gets stressed.\n\n**Step 3 — Diagnose the mental representation gap.** Ask the skill's core question: *what does an expert perceive here that the copilot has stopped forcing me to perceive?* An expert engineer sees the shape of a bug from the stack trace's structure; if you now paste every trace into a model, you never build that perception. An expert writer feels a paragraph's logical seam before writing the next sentence; if the draft is always AI-generated, you only ever edit, never compose. Name the perception you are no longer exercising — that is the atrophy site.\n\n**Step 4 — Design the repetition targeting the gap.** Build deliberate \"AI-off\" reps that isolate the endangered sub-skill and can be run dozens of times. Engineer: read a diff and predict the failure before running the tests; only then check. Writer: outline and draft a section from scratch, then use the model to critique structure — not to generate it. Analyst: write down the expected distribution of a dataset, then run the query and score the miss. Each rep must (a) trigger the perception the AI usually short-circuits, (b) produce in-session feedback (tests pass/fail, model critique, actual vs. predicted), and (c) be short enough to repeat many times.\n\n**Step 5 — Track representation progress, not output.** Shipped-PR count and word count will look fine or even rise *because* the copilot is carrying you — output is exactly the misleading metric here. Track perception instead: \"Am I predicting the bug before I run the code more often than last month? Am I anticipating the structural weakness in a draft before the model flags it?\" Progress is that your cold, AI-off prediction converges on the AI's answer — evidence your internal model, not the tool's, is improving.\n\n**Step 6 — Apply the stop-rule.** When the AI-off reps start feeling comfortable — when you reliably predict the model's output — that sub-skill has hardened and you should redesign toward a harder edge (a gnarlier codebase, an unfamiliar domain, a more ambiguous dataset). Comfort here is not success; it is the signal to move the practice frontier. And keep these reps inside protected windows: 1–4 genuine hours is Ericsson's ceiling, and the temptation under AI is to skip the hard reps entirely because the tool makes the comfortable path always available.\n\nThe mapped steps:\n1. Sub-skill defined precisely: the exact perception the copilot now performs first (predict-the-bug, compose-from-blank, hypothesize-before-query)\n2. Feedback mechanism constructed: attempt cold, then diff against the model's answer — seconds of latency\n3. Representation gap diagnosed: name the perception the AI has stopped forcing you to exercise\n4. Repetition designed to the gap: deliberate \"AI-off\" reps that trigger the perception and self-score in-session\n5. Progress tracked at representation level: convergence of your cold prediction on the correct answer, not shipped output\n6. Stop-rule applied: comfort/reliable prediction → move the frontier to a harder edge\n\nThe through-line to the timeless model: this is the same trap Ericsson identified in practitioners with decades of experience who plateaued — activity that has become automatic stops building representations. AI adoption simply accelerates and hides the transition into automaticity, because the tool, not the years, absorbs the reps. The counter is unchanged: deliberately re-inject the uncomfortable, feedback-rich repetition the routine used to supply by accident.\n\n*Sources: Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C., \"The Role of Deliberate Practice in the Acquisition of Expert Performance,\" *Psychological Review*, 100(3), 363–406, 1993, https://doi.org/10.1037/0033-295X.100.3.363. Ericsson, K.A. & Pool, R., *Peak: Secrets from the New Science of Expertise*, Houghton Mifflin Harcourt, 2016. On the 2024–2026 mainstreaming of AI coding/writing assistants: GitHub, \"Research: Quantifying GitHub Copilot's impact on developer productivity and happiness\" (2022) and the broadly reported enterprise adoption of Copilot, ChatGPT, and Claude across engineering and knowledge work through 2024–2025 (see e.g. Stack Overflow Developer Survey 2024, which reported a large majority of developers using or planning to use AI tools). Specific figures beyond these widely-reported findings are omitted to avoid overstatement.*\n\nFile v1.0.6:skill-card.md\n\n## Description:\n\nHelps an agent coach users through deliberate-practice design for overcoming skill plateaus, identifying expert mental representation gaps, and building fast feedback loops.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[deciqai](https://clawhub.ai/user/deciqai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal users, employees, and developers use this skill to design targeted practice programs when continued activity is not improving performance. It helps agents turn a plateau into a specific sub-skill, expert benchmark, feedback mechanism, repetition plan, and stop rule.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Practice plans may be ineffective if the user lacks a real expert benchmark or fast feedback signal.\n\nMitigation: Require a specific sub-skill, expert representation, feedback source, feedback latency, and stop rule before treating the plan as actionable.\n\nRisk: Research and AI-adoption claims could be over-applied outside the cited examples.\n\nMitigation: Use the cited sources as background guidance and validate important learning decisions with domain experts or observed performance evidence.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/deciqai/skills/deliberate-practice)\n- [deciqAI Deliberate Practice metadata](https://www.deciqai.com/s/deliberate-practice.json)\n- [deciqAI Deliberate Practice page](https://www.deciqai.com/c/deliberate-practice)\n- [Sources - deliberate-practice](references/sources.md)\n- [Berlin Violin Study example](examples/berlin-violin-study-1991-1993.md)\n- [Franklin Spectator Writing Method example](examples/franklin-spectator-writing-method.md)\n- [Keeping Skill Alive When AI Does the Reps example](examples/skill-atrophy-when-ai-does-the-reps-2024-2026.md)\n- [Ericsson, Krampe, and Tesch-Romer 1993](https://doi.org/10.1037/0033-295X.100.3.363)\n- [Stack Overflow Developer Survey 2024](https://survey.stackoverflow.co/2024/)\n- [GitHub Copilot productivity research](https://github.blog/2022-09-07-research-quantifying-github-copilots-impact-on-developer-productivity-and-happiness/)\n- [deciqAI knowledge-skills repository](https://github.com/deciqAI/knowledge-skills)\n\n## Skill Output:\n\n**Output Type(s):** [Text, Markdown, Guidance]\n\n**Output Format:** [Markdown practice design artifact with coaching questions and structured fields]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May include stepwise coaching questions with explicit wait points before advancing.]\n\n## Skill Version(s):\n\n1.0.6 (source: server release evidence)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nArchive v1.0.5: 7 files, 14473 bytes\n\nFiles: examples/berlin-violin-study-1991-1993.md (2172b), examples/franklin-spectator-writing-method.md (4709b), examples/skill-atrophy-when-ai-does-the-reps-2024-2026.md (6741b), references/sources.md (3019b), skill-card.md (2707b), SKILL.md (8140b), _meta.json (138b)\n\nFile v1.0.5:SKILL.md\n\n---\nname: deliberate-practice\ndescription: \"Activate when: user says 'I've been doing this for years but I'm not getting better'; someone suspects a skill plateau despite continued effort; designing a learning program for a high-performance outcome; an organization reports high training hours but low skill transfer.\n  Do NOT activate when: goal is execution of existing skills rather than acquiring new ones (use deep-work instead); there is no identifiable expert performance benchmark to target.\"\n---\n\n# Deliberate Practice\n\n## Overview\n\nMost people confuse repetition with learning — they accumulate years of experience and plateau. Once an activity becomes automatic, executing it no longer builds new neural architecture. Ericsson, Krampe & Tesch-Römer (1993) showed the predictive variable is not hours of doing but hours of *specifically deliberate practice* — targeted, uncomfortable, feedback-rich repetition designed to build mental representations.\n\n**Cross-skill composition:** Use `feedback-loops` first (audit your error signal); then `metacognition` (surface your current representation gap); use instead of `deep-work` when acquiring skills, not producing output; use alongside `cognitive-evolution-stages` for stage-aware practice design.\n\n---\n\n## When to Use\n\n**Trigger:** plateau despite experience; designing high-performance learning program; training hours high but skill transfer low; evaluating whether practice is building capability or maintaining it; skill atrophy or deskilling as AI copilots absorb the routine reps (AI adoption, AI hype, \"will AI make me worse at my craft\").\n**When NOT:** goal is execution not acquisition (use deep-work); no expert benchmark exists; bottleneck is motivational not representational.\n\n---\n\n## Coaching Novices (Adaptive Front Door)\n\n**Engine mode:** user has a concrete case → run The Process directly.\n**Coach mode:** user is unfamiliar or has no concrete case → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. Ask the plateau question: \"When you practice X, what does it feel like after 15 minutes — harder, the same, or easier?\" Comfort/easy = automatic = not building representations.\n2. Find the expert performance structure: \"Who is world-class at X? What do they perceive in the first 3 seconds that you don't?\" This locates the mental representation gap.\n3. Identify the discomfort zone: \"What part of practicing X makes you most want to stop?\" That is almost always where the gap lives.\n> **[WAIT — do not advance until user responds]**\n4. Design the smallest feedback loop: \"How would you know within 60 seconds whether a move was correct?\" Latency over 24h kills representation-building.\n> **[WAIT — do not advance until user responds]**\n5. Set the repetition target and stop-rule: \"How many reps of this specific discomfort can you sustain before concentration drops?\" (1–4 hours/day is Ericsson's ceiling.)\n> **[WAIT — do not advance until user responds]**\n\n---\n\n## The Process\n\n**Step 1 — Define the sub-skill with precision.** Not \"get better at X\" — specify the exact representational gap (e.g., \"detect when counterpart shifts from positional to interest-based\").\n**Step 2 — Find or construct the feedback mechanism.** Latency >24h breaks action-result association. Expert feedback > peer one level above > simulation with ground truth.\n**Step 3 — Diagnose the mental representation gap.** Ask: \"What does an expert *see* here that I don't?\" Not what they do — the doing follows from the seeing.\n**Step 4 — Design the repetition targeting the gap.** Must trigger the sub-skill, produce in-session feedback, and be executable at dozens–hundreds of reps per session.\n**Step 5 — Track representation progress, not output.** Output metrics lag by weeks. Track: \"Am I perceiving X earlier than before?\"\n**Step 6 — Apply the stop-rule.** Comfort = automaticity maintenance. Redesign to a harder sub-skill. End session when concentration drops.\n\n### Output: Practice Design Artifact\n\n```\nTarget sub-skill (precise): [specific representational gap]\nExpert mental representation: [what expert perceives that I currently don't]\nCurrent representation gap: [specific failure mode]\nFeedback — Source / Latency / Reliability:\nRepetition — Exercise / Volume / Duration / Frequency:\nProgress indicator (representation-level, not output): [what I will perceive by Week N]\nStop-rule triggers: comfortable → redesign; concentration drops → end; latency >24h → redesign\n```\n\n*→ Method in Action: [Berlin Violin Study (1991–1993)](examples/berlin-violin-study-1991-1993.md) · [Franklin's Spectator Method](examples/franklin-spectator-writing-method.md)*\n*→ 2026 lens: [Keeping Skill Alive When AI Does the Reps (2024–2026)](examples/skill-atrophy-when-ai-does-the-reps-2024-2026.md)*\n\n---\n\n## Practice Design Domain Packs\n\n**Medicine/Surgery:** sub-skill: laparoscopic tissue manipulation; feedback: simulator + debrief within 1h; rationalization to reject: \"I'll improve with more cases.\"\n**Writing:** sub-skill: eliminate nominalization in first-draft prose; feedback: rewrite published paragraphs vs original; rationalization: \"I write every day.\"\n**Investment:** sub-skill: identify customer concentration risk from footnotes in 20 min; feedback: 50-case retrospective library with outcomes.\n\nContribute packs via the deciqAI repo — requires sub-skill, expert representation, feedback latency, and common rationalization.\n\n---\n\n## Applying It Well\n\n- Target representations, not outcomes — ask \"What does the expert *perceive* that I don't?\"\n- Make feedback faster — redesign question: \"How do I get a reliable signal within 60 seconds?\"\n- Comfort signals time to redesign, not celebrate.\n- 1–4 genuine hours/day is Ericsson's hard ceiling; volume in degraded concentration reinforces errors.\n- Require expert think-alouds or annotated examples — you cannot design practice you cannot see.\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n---\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"I've been doing this for 10 years.\" | Duration is not deliberate practice. Years in automaticity = maintenance, not development. |\n| [D] \"I practice every day.\" | Comfortable daily repetition is automaticity reinforcement, not representation-building. |\n| [D] \"More cases/reps will help.\" | Only if structured to exceed current capability with rapid feedback. Otherwise more reps deepen the rut. |\n| [D] \"I can give myself feedback.\" | Self-feedback confirms what you already believe. External feedback from someone who sees the expert standard is required. |\n| [D] \"The discomfort means I'm doing it wrong.\" | Discomfort is the signal you are in deliberate practice. Comfort means automaticity. |\n| [D] \"My metrics are going up.\" | Output lags representation by weeks and is confounded by external factors. |\n| *→ Add [O] entries after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n---\n\n## Red Flags / Verification\n\n- Sessions feel comfortable — automaticity has absorbed the activity.\n- Feedback latency measured in days — action-result association cannot form.\n- Practitioner describes what expert *does* but not what they *perceive* — no representational target.\n- Practice volume cited as expertise without verifying hours were deliberate.\n- [ ] Sub-skill = specific representational gap; feedback latency <24h; expert representation identified.\n- [ ] Reps: dozens per session; stop-rule applied; progress tracked at representation level not output level.\n\n---\n\n*Part of **deciqAI Knowledge Skills** — 223 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. **See it run → https://www.deciqai.com/c/deliberate-practice** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.*\n\nFile v1.0.5:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"deliberate-practice\",\n  \"version\": \"1.0.5\",\n  \"publishedAt\": 1783679098498\n}\n\nFile v1.0.5:references/sources.md\n\n# Sources — deliberate-practice\n\n> *Primary sources for the [deliberate-practice](../SKILL.md) skill.*\n\n**Primary sources with verbatim quotes:**\n\n1. Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C. \"The Role of Deliberate Practice in the Acquisition of Expert Performance.\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n   > \"Deliberate practice includes activities that have been specially designed to improve the current level of performance. These activities demand full concentration and effort and are not inherently enjoyable.\"\n\n2. Ericsson, K.A. & Pool, R. *Peak: Secrets from the New Science of Expertise*. Houghton Mifflin Harcourt, 2016.\n   > \"Mental representations are what distinguish experts from novices... The main thing that sets experts apart from the rest of us is that their years of practice have changed the neural circuitry in their brains to produce highly specialized mental representations, which in turn make possible the incredible memory, pattern recognition, problem solving, and other sorts of advanced abilities needed to excel in their particular specialties.\"\n\n3. Franklin, B. *The Autobiography of Benjamin Franklin*. Part One, written 1771. Widely available; e.g. Project Gutenberg ebook #148.\n   > \"Then I compared my Spectator with the original, discovered some of my faults, and corrected them.\"\n\n**Contemporary context sources (2024–2026 AI-adoption example):**\n\n4. Stack Overflow. *2024 Developer Survey* (AI section). Reported that a large majority of professional developers were using or planning to use AI tools in their development workflow. https://survey.stackoverflow.co/2024/\n   - Used only to establish the durable, widely-reported fact that AI coding/writing assistants became mainstream in professional workflows by 2024–2025. No precise percentages are asserted in the example beyond \"a large majority.\"\n\n5. GitHub. \"Research: Quantifying GitHub Copilot's impact on developer productivity and happiness.\" GitHub Blog, 2022. https://github.blog/2022-09-07-research-quantifying-github-copilots-impact-on-developer-productivity-and-happiness/\n   - Cited as background evidence for Copilot's productivity effect and its adoption as a routine tool; the deskilling/atrophy risk in the example is a *prediction from Ericsson's model*, not a claim of a measured 2024–2026 effect.\n\n**What is NOT cited and why:**\n\n- Gladwell, M. *Outliers* (2008) is deliberately excluded. Gladwell popularized the 10,000-hour figure while dropping the \"deliberate\" qualifier — the central mechanism of Ericsson's finding. Citing Gladwell would reproduce the distortion this skill is designed to correct.\n- Chase & Simon (1973) \"Perception in Chess\" (*Cognitive Psychology*, 4(1), 55–81) is the foundational precursor study on expert mental representations in chess — relevant but not cited directly because Ericsson's 1993 paper synthesizes and extends this work and is the canonical source for deliberate practice specifically.\n\nFile v1.0.5:examples/berlin-violin-study-1991-1993.md\n\n# Method in Action: Berlin Violin Study (1991–1993)\n\n> *Example for the [deliberate-practice](../SKILL.md) skill.*\n\n**Source:** Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C., \"The Role of Deliberate Practice in the Acquisition of Expert Performance,\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n\nEricsson and colleagues studied violin students at the Musikhochschule (Academy of Music) in West Berlin. Teachers nominated students into three groups based on their assessment of students' potential: (1) the \"best\" violinists — those judged capable of international solo careers; (2) \"good\" violinists — talented but below the top group; (3) music teachers — students training to teach, not perform at elite level.\n\nAll three groups had begun playing at approximately age 5. By age 20:\n\n- **Best violinists** had accumulated an estimated **10,000 hours** of deliberate practice (defined specifically as practice designed to improve performance, not performance itself or music-related activities like theory or group rehearsal).\n- **Good violinists** had accumulated approximately **8,000 hours**.\n- **Music teachers** had accumulated approximately **4,000 hours**.\n\nThe critical methodological point — erased by Malcolm Gladwell's 2008 popularization — is that Ericsson carefully distinguished *deliberate practice* (specifically designed, uncomfortable, feedback-rich, coach-structured) from total music-related activity time. The predictive variable was not \"hours of playing\" but \"hours of specifically deliberate practice.\" The best violinists also rated deliberate practice as significantly less enjoyable than performance, yet they allocated more time to it — evidence that intrinsic enjoyment does not drive deliberate practice; external structure and long-term goal commitment do.\n\nThis study is the primary empirical foundation for understanding expert performance. It shows that the representation gap between good and best performers is not mysterious — it is the cumulative product of structured, targeted, uncomfortable repetition designed to build increasingly sophisticated internal models.\n\nFile v1.0.5:examples/franklin-spectator-writing-method.md\n\n# Method in Action: Benjamin Franklin's Spectator Writing Method (c. 1718–1723)\n\n> *Example for the [deliberate-practice](../SKILL.md) skill.*\n\nAs a teenage printer's apprentice in Boston, Benjamin Franklin engineered — two centuries before Ericsson named the mechanism — a complete deliberate practice system for prose writing, documented in detail in his *Autobiography*. It is the canonical historical case of self-designed practice in a domain with no available coach.\n\n**Step 1 — Define the sub-skill with precision.** Franklin did not resolve to \"write better.\" His father had compared his letters (from a written debate with his friend John Collins) against Collins's and identified specific deficits: Franklin's writing fell short in elegance of expression, in method, and in clarity. Franklin took these as three distinct representational gaps — word choice, arrangement of thoughts, and expression — and attacked each separately.\n\n**Step 2 — Find or construct the feedback mechanism.** With no writing tutor available, Franklin constructed a ground-truth comparator: an odd volume of *The Spectator*, the London periodical of Addison and Steele, whose prose he judged excellent. The original essay itself became the expert benchmark against which every attempt could be scored — feedback latency of minutes, not days.\n\n**Step 3 — Diagnose the mental representation gap.** The exercise structure forced the diagnosis. Franklin made short hints of the sentiment of each sentence in an essay, set them aside for a few days, then attempted to reconstruct the full essay from the hints in his own words. Comparing his reconstruction against Addison's original exposed exactly where his internal model of good prose diverged from the expert's — fault by fault, sentence by sentence.\n\n**Step 4 — Design the repetition targeting the gap.** When comparison revealed his stock of words was too small, Franklin redesigned the drill: he turned Spectator essays into verse and, after forgetting the originals, back into prose — because versification forces a continual search for words of different lengths and sounds to fit meter and rhyme. When the gap was arrangement rather than vocabulary, he jumbled his sentence hints into confusion, waited weeks, and then attempted to reorder them into the best structure before reconstructing — a drill isolating the method-of-thought sub-skill alone.\n\n**Step 5 — Track representation progress, not output.** Franklin's progress measure was perceptual, not productive: he tracked the faults he could now *discover* in his own reconstructions and correct. He records that he sometimes had the pleasure of judging that, in small particulars, he had improved on the original's method or language — evidence his internal representation of good prose had begun to match, and locally exceed, the benchmark.\n\n**Step 6 — Apply the stop-rule.** The escalation from plain reconstruction, to verse conversion, to jumbled-hint reordering is the stop-rule in action: each time an exercise stopped exposing faults, Franklin redesigned toward a harder sub-skill rather than accumulating comfortable reps. He also confined practice to protected windows — before work in the morning and after it at night — rather than diluting it across the day.\n\nThe outcome: Franklin became the most widely read prose stylist in colonial America, and his *Autobiography* passage became a standard exhibit in the modern science of expertise — Ericsson and Pool analyze it in *Peak* (2016) as a model of practice design under the constraint of no teacher.\n\nNote the counter to the \"I write every day\" rationalization: Franklin already spent his working days surrounded by text in the print shop. The day job never closed the gap — the designed, uncomfortable, feedback-rich exercises did.\n\nThe mapped steps:\n1. Sub-skill defined precisely: three named gaps (word stock, arrangement, expression), attacked separately\n2. Feedback mechanism constructed: the Spectator original as ground truth, comparison within the session\n3. Representation gap diagnosed: reconstruct-and-compare exposes divergence fault by fault\n4. Repetition designed to the gap: verse conversion for vocabulary; jumbled hints for structure\n5. Progress tracked at representation level: faults he could newly perceive, not pages produced\n6. Stop-rule applied: exercise redesigned harder whenever it stopped exposing faults\n\nPrimary source: Franklin, B. *The Autobiography of Benjamin Franklin* (Part One, written 1771; widely available, e.g. Project Gutenberg ebook #148). Analyzed in Ericsson, K.A. & Pool, R., *Peak: Secrets from the New Science of Expertise*, Houghton Mifflin Harcourt, 2016.\n\nFile v1.0.5:examples/skill-atrophy-when-ai-does-the-reps-2024-2026.md\n\n# Method in Action: Keeping Skill Alive When AI Does the Reps (2024–2026)\n\n> *Example for the [deliberate-practice](../SKILL.md) skill.*\n\nOver 2024–2025, AI coding assistants, writing copilots, and analysis tools moved from novelty to default, and by 2026 that shift was well underway. GitHub Copilot, ChatGPT, Claude, and similar tools became embedded in daily engineering, writing, and analyst workflows. This created a specific hazard that deliberate-practice theory predicts precisely: when a copilot handles the routine reps, the practitioner stops doing the very repetitions that once — almost accidentally — built and maintained their mental representations. The activity looks the same on the calendar (still \"coding,\" still \"writing\"), but the skill-building substrate has been quietly outsourced.\n\nThe failure mode is not laziness. It is that *offloading the routine also offloads the practice*. A junior analyst who once wrote every SQL query by hand built pattern recognition as a side effect; a copilot that autocompletes the query removes both the tedium and the representation-building. Below is how a deliberate engineer, writer, or analyst applies this skill to keep building expertise even as AI absorbs the routine work.\n\n**Step 1 — Define the sub-skill with precision.** Not \"stay sharp despite AI.\" Name the exact representational gap that the copilot is now filling on your behalf. For an engineer: \"reason about why a proposed implementation is wrong before running it — hold the data flow in my head rather than pasting the error back to the model.\" For a writer: \"structure an argument from a blank page — sequence claims and transitions without a draft to react to.\" For an analyst: \"form a hypothesis about what a dataset will show *before* querying it.\" The gap is always the perception step the AI now performs first.\n\n**Step 2 — Find or construct the feedback mechanism.** The AI itself is a fast ground-truth comparator if used deliberately rather than as a crutch. Attempt the task cold first; then ask the model for its solution and diff it against yours. Latency is seconds — far inside the 24-hour window where action-result association still forms. The critical discipline: generate your own answer *before* seeing the model's, or the comparison degrades into passive reading and the representation never gets stressed.\n\n**Step 3 — Diagnose the mental representation gap.** Ask the skill's core question: *what does an expert perceive here that the copilot has stopped forcing me to perceive?* An expert engineer sees the shape of a bug from the stack trace's structure; if you now paste every trace into a model, you never build that perception. An expert writer feels a paragraph's logical seam before writing the next sentence; if the draft is always AI-generated, you only ever edit, never compose. Name the perception you are no longer exercising — that is the atrophy site.\n\n**Step 4 — Design the repetition targeting the gap.** Build deliberate \"AI-off\" reps that isolate the endangered sub-skill and can be run dozens of times. Engineer: read a diff and predict the failure before running the tests; only then check. Writer: outline and draft a section from scratch, then use the model to critique structure — not to generate it. Analyst: write down the expected distribution of a dataset, then run the query and score the miss. Each rep must (a) trigger the perception the AI usually short-circuits, (b) produce in-session feedback (tests pass/fail, model critique, actual vs. predicted), and (c) be short enough to repeat many times.\n\n**Step 5 — Track representation progress, not output.** Shipped-PR count and word count will look fine or even rise *because* the copilot is carrying you — output is exactly the misleading metric here. Track perception instead: \"Am I predicting the bug before I run the code more often than last month? Am I anticipating the structural weakness in a draft before the model flags it?\" Progress is that your cold, AI-off prediction converges on the AI's answer — evidence your internal model, not the tool's, is improving.\n\n**Step 6 — Apply the stop-rule.** When the AI-off reps start feeling comfortable — when you reliably predict the model's output — that sub-skill has hardened and you should redesign toward a harder edge (a gnarlier codebase, an unfamiliar domain, a more ambiguous dataset). Comfort here is not success; it is the signal to move the practice frontier. And keep these reps inside protected windows: 1–4 genuine hours is Ericsson's ceiling, and the temptation under AI is to skip the hard reps entirely because the tool makes the comfortable path always available.\n\nThe mapped steps:\n1. Sub-skill defined precisely: the exact perception the copilot now performs first (predict-the-bug, compose-from-blank, hypothesize-before-query)\n2. Feedback mechanism constructed: attempt cold, then diff against the model's answer — seconds of latency\n3. Representation gap diagnosed: name the perception the AI has stopped forcing you to exercise\n4. Repetition designed to the gap: deliberate \"AI-off\" reps that trigger the perception and self-score in-session\n5. Progress tracked at representation level: convergence of your cold prediction on the correct answer, not shipped output\n6. Stop-rule applied: comfort/reliable prediction → move the frontier to a harder edge\n\nThe through-line to the timeless model: this is the same trap Ericsson identified in practitioners with decades of experience who plateaued — activity that has become automatic stops building representations. AI adoption simply accelerates and hides the transition into automaticity, because the tool, not the years, absorbs the reps. The counter is unchanged: deliberately re-inject the uncomfortable, feedback-rich repetition the routine used to supply by accident.\n\n*Sources: Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C., \"The Role of Deliberate Practice in the Acquisition of Expert Performance,\" *Psychological Review*, 100(3), 363–406, 1993, https://doi.org/10.1037/0033-295X.100.3.363. Ericsson, K.A. & Pool, R., *Peak: Secrets from the New Science of Expertise*, Houghton Mifflin Harcourt, 2016. On the 2024–2026 mainstreaming of AI coding/writing assistants: GitHub, \"Research: Quantifying GitHub Copilot's impact on developer productivity and happiness\" (2022) and the broadly reported enterprise adoption of Copilot, ChatGPT, and Claude across engineering and knowledge work through 2024–2025 (see e.g. Stack Overflow Developer Survey 2024, which reported a large majority of developers using or planning to use AI tools). Specific figures beyond these widely-reported findings are omitted to avoid overstatement.*\n\nFile v1.0.5:skill-card.md\n\n## Description: <br>\nHelps an agent coach users through deliberate practice design for skill plateaus, high-performance learning programs, and training-transfer problems. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nEmployees, external users, and developers use this skill when continued practice is not improving performance, when designing a high-performance learning program, or when assessing whether training is transferring into capability. The skill helps an agent identify representational gaps, design fast feedback loops, set repetition targets, and produce a practice design artifact. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Users may treat the skill's research-based coaching as professional medical, educational, workplace policy, or other regulated advice. <br>\nMitigation: Use the output as learning guidance and review domain-specific plans with qualified experts, instructors, supervisors, or applicable policy owners. <br>\nRisk: A practice plan can be misleading if the user lacks a real expert benchmark or rapid feedback loop. <br>\nMitigation: Apply the skill's verification checks before use: define a specific sub-skill, identify an expert representation, keep feedback latency under 24 hours, and redesign when practice becomes comfortable. <br>\n\n\n## Reference(s): <br>\n- [ClawHub Skill Page](https://clawhub.ai/deciqai/skills/deliberate-practice) <br>\n- [Sources - deliberate-practice](artifact/references/sources.md) <br>\n- [Ericsson, Krampe, and Tesch-Romer 1993](https://doi.org/10.1037/0033-295X.100.3.363) <br>\n- [Stack Overflow 2024 Developer Survey](https://survey.stackoverflow.co/2024/) <br>\n- [GitHub Copilot Productivity Research](https://github.blog/2022-09-07-research-quantifying-github-copilots-impact-on-developer-productivity-and-happiness/) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [text, markdown, guidance] <br>\n**Output Format:** [Markdown practice design artifact and step-by-step coaching prompts] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [May pause for user responses at [WAIT] steps; produces no executable code and requires no sensitive access.] <br>\n\n## Skill Version(s): <br>\n1.0.5 (source: server release metadata) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>\n\nArchive v1.0.4: 6 files, 10252 bytes\n\nFiles: examples/berlin-violin-study-1991-1993.md (2172b), examples/franklin-spectator-writing-method.md (4709b), references/sources.md (2022b), skill-card.md (2393b), SKILL.md (7878b), _meta.json (138b)\n\nFile v1.0.4:SKILL.md\n\n---\nname: deliberate-practice\ndescription: \"Activate when: user says 'I've been doing this for years but I'm not getting better'; someone suspects a skill plateau despite continued effort; designing a learning program for a high-performance outcome; an organization reports high training hours but low skill transfer.\n  Do NOT activate when: goal is execution of existing skills rather than acquiring new ones (use deep-work instead); there is no identifiable expert performance benchmark to target.\"\n---\n\n# Deliberate Practice\n\n## Overview\n\nMost people confuse repetition with learning — they accumulate years of experience and plateau. Once an activity becomes automatic, executing it no longer builds new neural architecture. Ericsson, Krampe & Tesch-Römer (1993) showed the predictive variable is not hours of doing but hours of *specifically deliberate practice* — targeted, uncomfortable, feedback-rich repetition designed to build mental representations.\n\n**Cross-skill composition:** Use `feedback-loops` first (audit your error signal); then `metacognition` (surface your current representation gap); use instead of `deep-work` when acquiring skills, not producing output; use alongside `cognitive-evolution-stages` for stage-aware practice design.\n\n---\n\n## When to Use\n\n**Trigger:** plateau despite experience; designing high-performance learning program; training hours high but skill transfer low; evaluating whether practice is building capability or maintaining it.\n**When NOT:** goal is execution not acquisition (use deep-work); no expert benchmark exists; bottleneck is motivational not representational.\n\n---\n\n## Coaching Novices (Adaptive Front Door)\n\n**Engine mode:** user has a concrete case → run The Process directly.\n**Coach mode:** user is unfamiliar or has no concrete case → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. Ask the plateau question: \"When you practice X, what does it feel like after 15 minutes — harder, the same, or easier?\" Comfort/easy = automatic = not building representations.\n2. Find the expert performance structure: \"Who is world-class at X? What do they perceive in the first 3 seconds that you don't?\" This locates the mental representation gap.\n3. Identify the discomfort zone: \"What part of practicing X makes you most want to stop?\" That is almost always where the gap lives.\n> **[WAIT — do not advance until user responds]**\n4. Design the smallest feedback loop: \"How would you know within 60 seconds whether a move was correct?\" Latency over 24h kills representation-building.\n> **[WAIT — do not advance until user responds]**\n5. Set the repetition target and stop-rule: \"How many reps of this specific discomfort can you sustain before concentration drops?\" (1–4 hours/day is Ericsson's ceiling.)\n> **[WAIT — do not advance until user responds]**\n\n---\n\n## The Process\n\n**Step 1 — Define the sub-skill with precision.** Not \"get better at X\" — specify the exact representational gap (e.g., \"detect when counterpart shifts from positional to interest-based\").\n**Step 2 — Find or construct the feedback mechanism.** Latency >24h breaks action-result association. Expert feedback > peer one level above > simulation with ground truth.\n**Step 3 — Diagnose the mental representation gap.** Ask: \"What does an expert *see* here that I don't?\" Not what they do — the doing follows from the seeing.\n**Step 4 — Design the repetition targeting the gap.** Must trigger the sub-skill, produce in-session feedback, and be executable at dozens–hundreds of reps per session.\n**Step 5 — Track representation progress, not output.** Output metrics lag by weeks. Track: \"Am I perceiving X earlier than before?\"\n**Step 6 — Apply the stop-rule.** Comfort = automaticity maintenance. Redesign to a harder sub-skill. End session when concentration drops.\n\n### Output: Practice Design Artifact\n\n```\nTarget sub-skill (precise): [specific representational gap]\nExpert mental representation: [what expert perceives that I currently don't]\nCurrent representation gap: [specific failure mode]\nFeedback — Source / Latency / Reliability:\nRepetition — Exercise / Volume / Duration / Frequency:\nProgress indicator (representation-level, not output): [what I will perceive by Week N]\nStop-rule triggers: comfortable → redesign; concentration drops → end; latency >24h → redesign\n```\n\n*→ Method in Action: [Berlin Violin Study (1991–1993)](examples/berlin-violin-study-1991-1993.md) · [Franklin's Spectator Method](examples/franklin-spectator-writing-method.md)*\n\n---\n\n## Practice Design Domain Packs\n\n**Medicine/Surgery:** sub-skill: laparoscopic tissue manipulation; feedback: simulator + debrief within 1h; rationalization to reject: \"I'll improve with more cases.\"\n**Writing:** sub-skill: eliminate nominalization in first-draft prose; feedback: rewrite published paragraphs vs original; rationalization: \"I write every day.\"\n**Investment:** sub-skill: identify customer concentration risk from footnotes in 20 min; feedback: 50-case retrospective library with outcomes.\n\nContribute packs via the deciqAI repo — requires sub-skill, expert representation, feedback latency, and common rationalization.\n\n---\n\n## Applying It Well\n\n- Target representations, not outcomes — ask \"What does the expert *perceive* that I don't?\"\n- Make feedback faster — redesign question: \"How do I get a reliable signal within 60 seconds?\"\n- Comfort signals time to redesign, not celebrate.\n- 1–4 genuine hours/day is Ericsson's hard ceiling; volume in degraded concentration reinforces errors.\n- Require expert think-alouds or annotated examples — you cannot design practice you cannot see.\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n---\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"I've been doing this for 10 years.\" | Duration is not deliberate practice. Years in automaticity = maintenance, not development. |\n| [D] \"I practice every day.\" | Comfortable daily repetition is automaticity reinforcement, not representation-building. |\n| [D] \"More cases/reps will help.\" | Only if structured to exceed current capability with rapid feedback. Otherwise more reps deepen the rut. |\n| [D] \"I can give myself feedback.\" | Self-feedback confirms what you already believe. External feedback from someone who sees the expert standard is required. |\n| [D] \"The discomfort means I'm doing it wrong.\" | Discomfort is the signal you are in deliberate practice. Comfort means automaticity. |\n| [D] \"My metrics are going up.\" | Output lags representation by weeks and is confounded by external factors. |\n| *→ Add [O] entries after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n---\n\n## Red Flags / Verification\n\n- Sessions feel comfortable — automaticity has absorbed the activity.\n- Feedback latency measured in days — action-result association cannot form.\n- Practitioner describes what expert *does* but not what they *perceive* — no representational target.\n- Practice volume cited as expertise without verifying hours were deliberate.\n- [ ] Sub-skill = specific representational gap; feedback latency <24h; expert representation identified.\n- [ ] Reps: dozens per session; stop-rule applied; progress tracked at representation level not output level.\n\n---\n\n*Part of **deciqAI Knowledge Skills** — 164 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. **See it run → https://www.deciqai.com/c/deliberate-practice** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.*\n\nFile v1.0.4:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"deliberate-practice\",\n  \"version\": \"1.0.4\",\n  \"publishedAt\": 1783508360733\n}\n\nFile v1.0.4:references/sources.md\n\n# Sources — deliberate-practice\n\n> *Primary sources for the [deliberate-practice](../SKILL.md) skill.*\n\n**Primary sources with verbatim quotes:**\n\n1. Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C. \"The Role of Deliberate Practice in the Acquisition of Expert Performance.\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n   > \"Deliberate practice includes activities that have been specially designed to improve the current level of performance. These activities demand full concentration and effort and are not inherently enjoyable.\"\n\n2. Ericsson, K.A. & Pool, R. *Peak: Secrets from the New Science of Expertise*. Houghton Mifflin Harcourt, 2016.\n   > \"Mental representations are what distinguish experts from novices... The main thing that sets experts apart from the rest of us is that their years of practice have changed the neural circuitry in their brains to produce highly specialized mental representations, which in turn make possible the incredible memory, pattern recognition, problem solving, and other sorts of advanced abilities needed to excel in their particular specialties.\"\n\n3. Franklin, B. *The Autobiography of Benjamin Franklin*. Part One, written 1771. Widely available; e.g. Project Gutenberg ebook #148.\n   > \"Then I compared my Spectator with the original, discovered some of my faults, and corrected them.\"\n\n**What is NOT cited and why:**\n\n- Gladwell, M. *Outliers* (2008) is deliberately excluded. Gladwell popularized the 10,000-hour figure while dropping the \"deliberate\" qualifier — the central mechanism of Ericsson's finding. Citing Gladwell would reproduce the distortion this skill is designed to correct.\n- Chase & Simon (1973) \"Perception in Chess\" (*Cognitive Psychology*, 4(1), 55–81) is the foundational precursor study on expert mental representations in chess — relevant but not cited directly because Ericsson's 1993 paper synthesizes and extends this work and is the canonical source for deliberate practice specifically.\n\nFile v1.0.4:examples/berlin-violin-study-1991-1993.md\n\n# Method in Action: Berlin Violin Study (1991–1993)\n\n> *Example for the [deliberate-practice](../SKILL.md) skill.*\n\n**Source:** Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C., \"The Role of Deliberate Practice in the Acquisition of Expert Performance,\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n\nEricsson and colleagues studied violin students at the Musikhochschule (Academy of Music) in West Berlin. Teachers nominated students into three groups based on their assessment of students' potential: (1) the \"best\" violinists — those judged capable of international solo careers; (2) \"good\" violinists — talented but below the top group; (3) music teachers — students training to teach, not perform at elite level.\n\nAll three groups had begun playing at approximately age 5. By age 20:\n\n- **Best violinists** had accumulated an estimated **10,000 hours** of deliberate practice (defined specifically as practice designed to improve performance, not performance itself or music-related activities like theory or group rehearsal).\n- **Good violinists** had accumulated approximately **8,000 hours**.\n- **Music teachers** had accumulated approximately **4,000 hours**.\n\nThe critical methodological point — erased by Malcolm Gladwell's 2008 popularization — is that Ericsson carefully distinguished *deliberate practice* (specifically designed, uncomfortable, feedback-rich, coach-structured) from total music-related activity time. The predictive variable was not \"hours of playing\" but \"hours of specifically deliberate practice.\" The best violinists also rated deliberate practice as significantly less enjoyable than performance, yet they allocated more time to it — evidence that intrinsic enjoyment does not drive deliberate practice; external structure and long-term goal commitment do.\n\nThis study is the primary empirical foundation for understanding expert performance. It shows that the representation gap between good and best performers is not mysterious — it is the cumulative product of structured, targeted, uncomfortable repetition designed to build increasingly sophisticated internal models.\n\nFile v1.0.4:examples/franklin-spectator-writing-method.md\n\n# Method in Action: Benjamin Franklin's Spectator Writing Method (c. 1718–1723)\n\n> *Example for the [deliberate-practice](../SKILL.md) skill.*\n\nAs a teenage printer's apprentice in Boston, Benjamin Franklin engineered — two centuries before Ericsson named the mechanism — a complete deliberate practice system for prose writing, documented in detail in his *Autobiography*. It is the canonical historical case of self-designed practice in a domain with no available coach.\n\n**Step 1 — Define the sub-skill with precision.** Franklin did not resolve to \"write better.\" His father had compared his letters (from a written debate with his friend John Collins) against Collins's and identified specific deficits: Franklin's writing fell short in elegance of expression, in method, and in clarity. Franklin took these as three distinct representational gaps — word choice, arrangement of thoughts, and expression — and attacked each separately.\n\n**Step 2 — Find or construct the feedback mechanism.** With no writing tutor available, Franklin constructed a ground-truth comparator: an odd volume of *The Spectator*, the London periodical of Addison and Steele, whose prose he judged excellent. The original essay itself became the expert benchmark against which every attempt could be scored — feedback latency of minutes, not days.\n\n**Step 3 — Diagnose the mental representation gap.** The exercise structure forced the diagnosis. Franklin made short hints of the sentiment of each sentence in an essay, set them aside for a few days, then attempted to reconstruct the full essay from the hints in his own words. Comparing his reconstruction against Addison's original exposed exactly where his internal model of good prose diverged from the expert's — fault by fault, sentence by sentence.\n\n**Step 4 — Design the repetition targeting the gap.** When comparison revealed his stock of words was too small, Franklin redesigned the drill: he turned Spectator essays into verse and, after forgetting the originals, back into prose — because versification forces a continual search for words of different lengths and sounds to fit meter and rhyme. When the gap was arrangement rather than vocabulary, he jumbled his sentence hints into confusion, waited weeks, and then attempted to reorder them into the best structure before reconstructing — a drill isolating the method-of-thought sub-skill alone.\n\n**Step 5 — Track representation progress, not output.** Franklin's progress measure was perceptual, not productive: he tracked the faults he could now *discover* in his own reconstructions and correct. He records that he sometimes had the pleasure of judging that, in small particulars, he had improved on the original's method or language — evidence his internal representation of good prose had begun to match, and locally exceed, the benchmark.\n\n**Step 6 — Apply the stop-rule.** The escalation from plain reconstruction, to verse conversion, to jumbled-hint reordering is the stop-rule in action: each time an exercise stopped exposing faults, Franklin redesigned toward a harder sub-skill rather than accumulating comfortable reps. He also confined practice to protected windows — before work in the morning and after it at night — rather than diluting it across the day.\n\nThe outcome: Franklin became the most widely read prose stylist in colonial America, and his *Autobiography* passage became a standard exhibit in the modern science of expertise — Ericsson and Pool analyze it in *Peak* (2016) as a model of practice design under the constraint of no teacher.\n\nNote the counter to the \"I write every day\" rationalization: Franklin already spent his working days surrounded by text in the print shop. The day job never closed the gap — the designed, uncomfortable, feedback-rich exercises did.\n\nThe mapped steps:\n1. Sub-skill defined precisely: three named gaps (word stock, arrangement, expression), attacked separately\n2. Feedback mechanism constructed: the Spectator original as ground truth, comparison within the session\n3. Representation gap diagnosed: reconstruct-and-compare exposes divergence fault by fault\n4. Repetition designed to the gap: verse conversion for vocabulary; jumbled hints for structure\n5. Progress tracked at representation level: faults he could newly perceive, not pages produced\n6. Stop-rule applied: exercise redesigned harder whenever it stopped exposing faults\n\nPrimary source: Franklin, B. *The Autobiography of Benjamin Franklin* (Part One, written 1771; widely available, e.g. Project Gutenberg ebook #148). Analyzed in Ericsson, K.A. & Pool, R., *Peak: Secrets from the New Science of Expertise*, Houghton Mifflin Harcourt, 2016.\n\nFile v1.0.4:skill-card.md\n\n## Description: <br>\nHelps agents design deliberate-practice plans for skill plateaus by defining representational gaps, fast feedback loops, targeted repetitions, progress indicators, and stop rules. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nEmployees, external users, and agent builders use this skill to turn vague improvement goals into structured practice designs when continued experience is no longer improving performance. It is intended for coaching conversations where an expert benchmark, a specific representation gap, and a fast feedback mechanism can be identified. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Users may share private business or personal details while discussing skill gaps and practice design. <br>\nMitigation: Avoid entering sensitive details unless they are appropriate for the coaching context. <br>\nRisk: The skill asks reflective questions and may pause for answers, which can surprise users expecting immediate task execution. <br>\nMitigation: Use it in coaching or learning-design sessions where iterative questioning is expected. <br>\n\n\n## Reference(s): <br>\n- [Sources - deliberate-practice](references/sources.md) <br>\n- [Berlin Violin Study (1991-1993)](examples/berlin-violin-study-1991-1993.md) <br>\n- [Franklin's Spectator Writing Method](examples/franklin-spectator-writing-method.md) <br>\n- [Ericsson, Krampe, and Tesch-Romer 1993](https://doi.org/10.1037/0033-295X.100.3.363) <br>\n- [ClawHub skill page](https://clawhub.ai/deciqai/skills/deliberate-practice) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [text, markdown, guidance] <br>\n**Output Format:** [Markdown coaching questions and a structured practice design artifact] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [May pause for user answers before advancing through coaching steps.] <br>\n\n## Skill Version(s): <br>\n1.0.4 (source: server release evidence) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>\n\nArchive v1.0.3: 6 files, 10313 bytes\n\nFiles: examples/berlin-violin-study-1991-1993.md (2172b), examples/franklin-spectator-writing-method.md (4709b), references/sources.md (2022b), skill-card.md (2411b), SKILL.md (7987b), _meta.json (138b)\n\nFile v1.0.3:SKILL.md\n\n---\nname: deliberate-practice\ndescription: \"Activate when: user says 'I've been doing this for years but I'm not getting better'; someone suspects a skill plateau despite continued effort; designing a learning program for a high-performance outcome; an organization reports high training hours but low skill transfer.\n  Do NOT activate when: goal is execution of existing skills rather than acquiring new ones (use deep-work instead); there is no identifiable expert performance benchmark to target.\"\n---\n\n# Deliberate Practice\n\n## Overview\n\nMost people confuse repetition with learning — they accumulate years of experience and plateau. Once an activity becomes automatic, executing it no longer builds new neural architecture. Ericsson, Krampe & Tesch-Römer (1993) showed the predictive variable is not hours of doing but hours of *specifically deliberate practice* — targeted, uncomfortable, feedback-rich repetition designed to build mental representations.\n\n**Cross-skill composition:** Use `feedback-loops` first (audit your error signal); then `metacognition` (surface your current representation gap); use instead of `deep-work` when acquiring skills, not producing output; use alongside `cognitive-evolution-stages` for stage-aware practice design.\n\n---\n\n## When to Use\n\n**Trigger:** plateau despite experience; designing high-performance learning program; training hours high but skill transfer low; evaluating whether practice is building capability or maintaining it.\n**When NOT:** goal is execution not acquisition (use deep-work); no expert benchmark exists; bottleneck is motivational not representational.\n\n---\n\n## Coaching Novices (Adaptive Front Door)\n\n**Engine mode:** user has a concrete case → run The Process directly.\n**Coach mode:** user is unfamiliar or has no concrete case → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. Ask the plateau question: \"When you practice X, what does it feel like after 15 minutes — harder, the same, or easier?\" Comfort/easy = automatic = not building representations.\n2. Find the expert performance structure: \"Who is world-class at X? What do they perceive in the first 3 seconds that you don't?\" This locates the mental representation gap.\n3. Identify the discomfort zone: \"What part of practicing X makes you most want to stop?\" That is almost always where the gap lives.\n> **[WAIT — do not advance until user responds]**\n4. Design the smallest feedback loop: \"How would you know within 60 seconds whether a move was correct?\" Latency over 24h kills representation-building.\n> **[WAIT — do not advance until user responds]**\n5. Set the repetition target and stop-rule: \"How many reps of this specific discomfort can you sustain before concentration drops?\" (1–4 hours/day is Ericsson's ceiling.)\n> **[WAIT — do not advance until user responds]**\n\n---\n\n## The Process\n\n**Step 1 — Define the sub-skill with precision.** Not \"get better at X\" — specify the exact representational gap (e.g., \"detect when counterpart shifts from positional to interest-based\").\n**Step 2 — Find or construct the feedback mechanism.** Latency >24h breaks action-result association. Expert feedback > peer one level above > simulation with ground truth.\n**Step 3 — Diagnose the mental representation gap.** Ask: \"What does an expert *see* here that I don't?\" Not what they do — the doing follows from the seeing.\n**Step 4 — Design the repetition targeting the gap.** Must trigger the sub-skill, produce in-session feedback, and be executable at dozens–hundreds of reps per session.\n**Step 5 — Track representation progress, not output.** Output metrics lag by weeks. Track: \"Am I perceiving X earlier than before?\"\n**Step 6 — Apply the stop-rule.** Comfort = automaticity maintenance. Redesign to a harder sub-skill. End session when concentration drops.\n\n### Output: Practice Design Artifact\n\n```\nTarget sub-skill (precise): [specific representational gap]\nExpert mental representation: [what expert perceives that I currently don't]\nCurrent representation gap: [specific failure mode]\nFeedback — Source / Latency / Reliability:\nRepetition — Exercise / Volume / Duration / Frequency:\nProgress indicator (representation-level, not output): [what I will perceive by Week N]\nStop-rule triggers: comfortable → redesign; concentration drops → end; latency >24h → redesign\n```\n\n*→ Method in Action: [Berlin Violin Study (1991–1993)](examples/berlin-violin-study-1991-1993.md) · [Franklin's Spectator Method](examples/franklin-spectator-writing-method.md)*\n\n---\n\n## Practice Design Domain Packs\n\n**Medicine/Surgery:** sub-skill: laparoscopic tissue manipulation; feedback: simulator + debrief within 1h; rationalization to reject: \"I'll improve with more cases.\"\n**Writing:** sub-skill: eliminate nominalization in first-draft prose; feedback: rewrite published paragraphs vs original; rationalization: \"I write every day.\"\n**Investment:** sub-skill: identify customer concentration risk from footnotes in 20 min; feedback: 50-case retrospective library with outcomes.\n\nContribute packs via the deciqAI repo — requires sub-skill, expert representation, feedback latency, and common rationalization.\n\n---\n\n## Applying It Well\n\n- Target representations, not outcomes — ask \"What does the expert *perceive* that I don't?\"\n- Make feedback faster — redesign question: \"How do I get a reliable signal within 60 seconds?\"\n- Comfort signals time to redesign, not celebrate.\n- 1–4 genuine hours/day is Ericsson's hard ceiling; volume in degraded concentration reinforces errors.\n- Require expert think-alouds or annotated examples — you cannot design practice you cannot see.\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n---\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"I've been doing this for 10 years.\" | Duration is not deliberate practice. Years in automaticity = maintenance, not development. |\n| [D] \"I practice every day.\" | Comfortable daily repetition is automaticity reinforcement, not representation-building. |\n| [D] \"More cases/reps will help.\" | Only if structured to exceed current capability with rapid feedback. Otherwise more reps deepen the rut. |\n| [D] \"I can give myself feedback.\" | Self-feedback confirms what you already believe. External feedback from someone who sees the expert standard is required. |\n| [D] \"The discomfort means I'm doing it wrong.\" | Discomfort is the signal you are in deliberate practice. Comfort means automaticity. |\n| [D] \"My metrics are going up.\" | Output lags representation by weeks and is confounded by external factors. |\n| *→ Add [O] entries after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n---\n\n## Red Flags / Verification\n\n- Sessions feel comfortable — automaticity has absorbed the activity.\n- Feedback latency measured in days — action-result association cannot form.\n- Practitioner describes what expert *does* but not what they *perceive* — no representational target.\n- Practice volume cited as expertise without verifying hours were deliberate.\n- [ ] Sub-skill = specific representational gap; feedback latency <24h; expert representation identified.\n- [ ] Reps: dozens per session; stop-rule applied; progress tracked at representation level not output level.\n\n---\n\n*Part of **deciqAI Knowledge Skills** — 163 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. **See it run → https://www.deciqai.com/skills/deliberate-practice?utm_source=clawhub&utm_medium=marketplace&utm_campaign=knowledge-skills&utm_content=deliberate-practice** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.*\n\nFile v1.0.3:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"deliberate-practice\",\n  \"version\": \"1.0.3\",\n  \"publishedAt\": 1783482691680\n}\n\nFile v1.0.3:references/sources.md\n\n# Sources — deliberate-practice\n\n> *Primary sources for the [deliberate-practice](../SKILL.md) skill.*\n\n**Primary sources with verbatim quotes:**\n\n1. Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C. \"The Role of Deliberate Practice in the Acquisition of Expert Performance.\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n   > \"Deliberate practice includes activities that have been specially designed to improve the current level of performance. These activities demand full concentration and effort and are not inherently enjoyable.\"\n\n2. Ericsson, K.A. & Pool, R. *Peak: Secrets from the New Science of Expertise*. Houghton Mifflin Harcourt, 2016.\n   > \"Mental representations are what distinguish experts from novices... The main thing that sets experts apart from the rest of us is that their years of practice have changed the neural circuitry in their brains to produce highly specialized mental representations, which in turn make possible the incredible memory, pattern recognition, problem solving, and other sorts of advanced abilities needed to excel in their particular specialties.\"\n\n3. Franklin, B. *The Autobiography of Benjamin Franklin*. Part One, written 1771. Widely available; e.g. Project Gutenberg ebook #148.\n   > \"Then I compared my Spectator with the original, discovered some of my faults, and corrected them.\"\n\n**What is NOT cited and why:**\n\n- Gladwell, M. *Outliers* (2008) is deliberately excluded. Gladwell popularized the 10,000-hour figure while dropping the \"deliberate\" qualifier — the central mechanism of Ericsson's finding. Citing Gladwell would reproduce the distortion this skill is designed to correct.\n- Chase & Simon (1973) \"Perception in Chess\" (*Cognitive Psychology*, 4(1), 55–81) is the foundational precursor study on expert mental representations in chess — relevant but not cited directly because Ericsson's 1993 paper synthesizes and extends this work and is the canonical source for deliberate practice specifically.\n\nFile v1.0.3:examples/berlin-violin-study-1991-1993.md\n\n# Method in Action: Berlin Violin Study (1991–1993)\n\n> *Example for the [deliberate-practice](../SKILL.md) skill.*\n\n**Source:** Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C., \"The Role of Deliberate Practice in the Acquisition of Expert Performance,\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n\nEricsson and colleagues studied violin students at the Musikhochschule (Academy of Music) in West Berlin. Teachers nominated students into three groups based on their assessment of students' potential: (1) the \"best\" violinists — those judged capable of international solo careers; (2) \"good\" violinists — talented but below the top group; (3) music teachers — students training to teach, not perform at elite level.\n\nAll three groups had begun playing at approximately age 5. By age 20:\n\n- **Best violinists** had accumulated an estimated **10,000 hours** of deliberate practice (defined specifically as practice designed to improve performance, not performance itself or music-related activities like theory or group rehearsal).\n- **Good violinists** had accumulated approximately **8,000 hours**.\n- **Music teachers** had accumulated approximately **4,000 hours**.\n\nThe critical methodological point — erased by Malcolm Gladwell's 2008 popularization — is that Ericsson carefully distinguished *deliberate practice* (specifically designed, uncomfortable, feedback-rich, coach-structured) from total music-related activity time. The predictive variable was not \"hours of playing\" but \"hours of specifically deliberate practice.\" The best violinists also rated deliberate practice as significantly less enjoyable than performance, yet they allocated more time to it — evidence that intrinsic enjoyment does not drive deliberate practice; external structure and long-term goal commitment do.\n\nThis study is the primary empirical foundation for understanding expert performance. It shows that the representation gap between good and best performers is not mysterious — it is the cumulative product of structured, targeted, uncomfortable repetition designed to build increasingly sophisticated internal models.\n\nFile v1.0.3:examples/franklin-spectator-writing-method.md\n\n# Method in Action: Benjamin Franklin's Spectator Writing Method (c. 1718–1723)\n\n> *Example for the [deliberate-practice](../SKILL.md) skill.*\n\nAs a teenage printer's apprentice in Boston, Benjamin Franklin engineered — two centuries before Ericsson named the mechanism — a complete deliberate practice system for prose writing, documented in detail in his *Autobiography*. It is the canonical historical case of self-designed practice in a domain with no available coach.\n\n**Step 1 — Define the sub-skill with precision.** Franklin did not resolve to \"write better.\" His father had compared his letters (from a written debate with his friend John Collins) against Collins's and identified specific deficits: Franklin's writing fell short in elegance of expression, in method, and in clarity. Franklin took these as three distinct representational gaps — word choice, arrangement of thoughts, and expression — and attacked each separately.\n\n**Step 2 — Find or construct the feedback mechanism.** With no writing tutor available, Franklin constructed a ground-truth comparator: an odd volume of *The Spectator*, the London periodical of Addison and Steele, whose prose he judged excellent. The original essay itself became the expert benchmark against which every attempt could be scored — feedback latency of minutes, not days.\n\n**Step 3 — Diagnose the mental representation gap.** The exercise structure forced the diagnosis. Franklin made short hints of the sentiment of each sentence in an essay, set them aside for a few days, then attempted to reconstruct the full essay from the hints in his own words. Comparing his reconstruction against Addison's original exposed exactly where his internal model of good prose diverged from the expert's — fault by fault, sentence by sentence.\n\n**Step 4 — Design the repetition targeting the gap.** When comparison revealed his stock of words was too small, Franklin redesigned the drill: he turned Spectator essays into verse and, after forgetting the originals, back into prose — because versification forces a continual search for words of different lengths and sounds to fit meter and rhyme. When the gap was arrangement rather than vocabulary, he jumbled his sentence hints into confusion, waited weeks, and then attempted to reorder them into the best structure before reconstructing — a drill isolating the method-of-thought sub-skill alone.\n\n**Step 5 — Track representation progress, not output.** Franklin's progress measure was perceptual, not productive: he tracked the faults he could now *discover* in his own reconstructions and correct. He records that he sometimes had the pleasure of judging that, in small particulars, he had improved on the original's method or language — evidence his internal representation of good prose had begun to match, and locally exceed, the benchmark.\n\n**Step 6 — Apply the stop-rule.** The escalation from plain reconstruction, to verse conversion, to jumbled-hint reordering is the stop-rule in action: each time an exercise stopped exposing faults, Franklin redesigned toward a harder sub-skill rather than accumulating comfortable reps. He also confined practice to protected windows — before work in the morning and after it at night — rather than diluting it across the day.\n\nThe outcome: Franklin became the most widely read prose stylist in colonial America, and his *Autobiography* passage became a standard exhibit in the modern science of expertise — Ericsson and Pool analyze it in *Peak* (2016) as a model of practice design under the constraint of no teacher.\n\nNote the counter to the \"I write every day\" rationalization: Franklin already spent his working days surrounded by text in the print shop. The day job never closed the gap — the designed, uncomfortable, feedback-rich exercises did.\n\nThe mapped steps:\n1. Sub-skill defined precisely: three named gaps (word stock, arrangement, expression), attacked separately\n2. Feedback mechanism constructed: the Spectator original as ground truth, comparison within the session\n3. Representation gap diagnosed: reconstruct-and-compare exposes divergence fault by fault\n4. Repetition designed to the gap: verse conversion for vocabulary; jumbled hints for structure\n5. Progress tracked at representation level: faults he could newly perceive, not pages produced\n6. Stop-rule applied: exercise redesigned harder whenever it stopped exposing faults\n\nPrimary source: Franklin, B. *The Autobiography of Benjamin Franklin* (Part One, written 1771; widely available, e.g. Project Gutenberg ebook #148). Analyzed in Ericsson, K.A. & Pool, R., *Peak: Secrets from the New Science of Expertise*, Houghton Mifflin Harcourt, 2016.\n\nFile v1.0.3:skill-card.md\n\n## Description: <br>\nDeliberate Practice helps an agent diagnose skill plateaus and design feedback-rich practice routines that target expert mental representations rather than repetition volume. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nEmployees, external users, coaches, and developers use this skill to turn vague skill-improvement goals into structured deliberate-practice plans with expert benchmarks, fast feedback, targeted repetitions, and stop rules. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Research and coaching claims may be over-applied without checking whether the user's domain has a clear expert benchmark and reliable feedback signal. <br>\nMitigation: Treat the skill's claims as guidance to evaluate, and require an expert benchmark, rapid feedback loop, and human judgment before using it for consequential training decisions. <br>\nRisk: The artifact includes external deciqAI, GitHub, and DOI links. <br>\nMitigation: Review external destinations before relying on them or sharing information through them. <br>\n\n\n## Reference(s): <br>\n- [Primary Sources](references/sources.md) <br>\n- [Berlin Violin Study Example](examples/berlin-violin-study-1991-1993.md) <br>\n- [Franklin's Spectator Method Example](examples/franklin-spectator-writing-method.md) <br>\n- [The Role of Deliberate Practice in the Acquisition of Expert Performance](https://doi.org/10.1037/0033-295X.100.3.363) <br>\n- [ClawHub Skill Listing](https://clawhub.ai/deciqai/skills/deliberate-practice) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [Guidance, Markdown, Text] <br>\n**Output Format:** [Markdown coaching guidance and a structured practice design artifact] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [May pause at WAIT steps for user input; does not run code, access private data, or modify the environment.] <br>\n\n## Skill Version(s): <br>\n1.0.3 (source: evidence.release.version) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>\n\nArchive v1.0.2: 5 files, 7771 bytes\n\nFiles: examples/berlin-violin-study-1991-1993.md (2172b), references/sources.md (1781b), skill-card.md (2529b), SKILL.md (7907b), _meta.json (138b)\n\nFile v1.0.2:SKILL.md\n\n---\nname: deliberate-practice\ndescription: \"Activate when: user says 'I've been doing this for years but I'm not getting better'; someone suspects a skill plateau despite continued effort; designing a learning program for a high-performance outcome; an organization reports high training hours but low skill transfer.\n  Do NOT activate when: goal is execution of existing skills rather than acquiring new ones (use deep-work instead); there is no identifiable expert performance benchmark to target.\"\n---\n\n# Deliberate Practice\n\n## Overview\n\nMost people confuse repetition with learning — they accumulate years of experience and plateau. Once an activity becomes automatic, executing it no longer builds new neural architecture. Ericsson, Krampe & Tesch-Römer (1993) showed the predictive variable is not hours of doing but hours of *specifically deliberate practice* — targeted, uncomfortable, feedback-rich repetition designed to build mental representations.\n\n**Cross-skill composition:** Use `feedback-loops` first (audit your error signal); then `metacognition` (surface your current representation gap); use instead of `deep-work` when acquiring skills, not producing output; use alongside `cognitive-evolution-stages` for stage-aware practice design.\n\n---\n\n## When to Use\n\n**Trigger:** plateau despite experience; designing high-performance learning program; training hours high but skill transfer low; evaluating whether practice is building capability or maintaining it.\n**When NOT:** goal is execution not acquisition (use deep-work); no expert benchmark exists; bottleneck is motivational not representational.\n\n---\n\n## Coaching Novices (Adaptive Front Door)\n\n**Engine mode:** user has a concrete case → run The Process directly.\n**Coach mode:** user is unfamiliar or has no concrete case → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. Ask the plateau question: \"When you practice X, what does it feel like after 15 minutes — harder, the same, or easier?\" Comfort/easy = automatic = not building representations.\n2. Find the expert performance structure: \"Who is world-class at X? What do they perceive in the first 3 seconds that you don't?\" This locates the mental representation gap.\n3. Identify the discomfort zone: \"What part of practicing X makes you most want to stop?\" That is almost always where the gap lives.\n> **[WAIT — do not advance until user responds]**\n4. Design the smallest feedback loop: \"How would you know within 60 seconds whether a move was correct?\" Latency over 24h kills representation-building.\n> **[WAIT — do not advance until user responds]**\n5. Set the repetition target and stop-rule: \"How many reps of this specific discomfort can you sustain before concentration drops?\" (1–4 hours/day is Ericsson's ceiling.)\n> **[WAIT — do not advance until user responds]**\n\n---\n\n## The Process\n\n**Step 1 — Define the sub-skill with precision.** Not \"get better at X\" — specify the exact representational gap (e.g., \"detect when counterpart shifts from positional to interest-based\").\n**Step 2 — Find or construct the feedback mechanism.** Latency >24h breaks action-result association. Expert feedback > peer one level above > simulation with ground truth.\n**Step 3 — Diagnose the mental representation gap.** Ask: \"What does an expert *see* here that I don't?\" Not what they do — the doing follows from the seeing.\n**Step 4 — Design the repetition targeting the gap.** Must trigger the sub-skill, produce in-session feedback, and be executable at dozens–hundreds of reps per session.\n**Step 5 — Track representation progress, not output.** Output metrics lag by weeks. Track: \"Am I perceiving X earlier than before?\"\n**Step 6 — Apply the stop-rule.** Comfort = automaticity maintenance. Redesign to a harder sub-skill. End session when concentration drops.\n\n### Output: Practice Design Artifact\n\n```\nTarget sub-skill (precise): [specific representational gap]\nExpert mental representation: [what expert perceives that I currently don't]\nCurrent representation gap: [specific failure mode]\nFeedback — Source / Latency / Reliability:\nRepetition — Exercise / Volume / Duration / Frequency:\nProgress indicator (representation-level, not output): [what I will perceive by Week N]\nStop-rule triggers: comfortable → redesign; concentration drops → end; latency >24h → redesign\n```\n\n*→ Method in Action: [Berlin Violin Study (1991–1993)](examples/berlin-violin-study-1991-1993.md)*\n\n---\n\n## Practice Design Domain Packs\n\n**Medicine/Surgery:** sub-skill: laparoscopic tissue manipulation; feedback: simulator + debrief within 1h; rationalization to reject: \"I'll improve with more cases.\"\n**Writing:** sub-skill: eliminate nominalization in first-draft prose; feedback: rewrite published paragraphs vs original; rationalization: \"I write every day.\"\n**Investment:** sub-skill: identify customer concentration risk from footnotes in 20 min; feedback: 50-case retrospective library with outcomes.\n\nContribute packs via the deciqAI repo — requires sub-skill, expert representation, feedback latency, and common rationalization.\n\n---\n\n## Applying It Well\n\n- Target representations, not outcomes — ask \"What does the expert *perceive* that I don't?\"\n- Make feedback faster — redesign question: \"How do I get a reliable signal within 60 seconds?\"\n- Comfort signals time to redesign, not celebrate.\n- 1–4 genuine hours/day is Ericsson's hard ceiling; volume in degraded concentration reinforces errors.\n- Require expert think-alouds or annotated examples — you cannot design practice you cannot see.\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n---\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"I've been doing this for 10 years.\" | Duration is not deliberate practice. Years in automaticity = maintenance, not development. |\n| [D] \"I practice every day.\" | Comfortable daily repetition is automaticity reinforcement, not representation-building. |\n| [D] \"More cases/reps will help.\" | Only if structured to exceed current capability with rapid feedback. Otherwise more reps deepen the rut. |\n| [D] \"I can give myself feedback.\" | Self-feedback confirms what you already believe. External feedback from someone who sees the expert standard is required. |\n| [D] \"The discomfort means I'm doing it wrong.\" | Discomfort is the signal you are in deliberate practice. Comfort means automaticity. |\n| [D] \"My metrics are going up.\" | Output lags representation by weeks and is confounded by external factors. |\n| *→ Add [O] entries after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n---\n\n## Red Flags / Verification\n\n- Sessions feel comfortable — automaticity has absorbed the activity.\n- Feedback latency measured in days — action-result association cannot form.\n- Practitioner describes what expert *does* but not what they *perceive* — no representational target.\n- Practice volume cited as expertise without verifying hours were deliberate.\n- [ ] Sub-skill = specific representational gap; feedback latency <24h; expert representation identified.\n- [ ] Reps: dozens per session; stop-rule applied; progress tracked at representation level not output level.\n\n---\n\n*Part of **deciqAI Knowledge Skills** — 163 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. **See it run → https://www.deciqai.com/skills/deliberate-practice?utm_source=clawhub&utm_medium=marketplace&utm_campaign=knowledge-skills&utm_content=deliberate-practice** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.*\n\nFile v1.0.2:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"deliberate-practice\",\n  \"version\": \"1.0.2\",\n  \"publishedAt\": 1783471455773\n}\n\nFile v1.0.2:references/sources.md\n\n# Sources — deliberate-practice\n\n> *Primary sources for the [deliberate-practice](../SKILL.md) skill.*\n\n**Primary sources with verbatim quotes:**\n\n1. Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C. \"The Role of Deliberate Practice in the Acquisition of Expert Performance.\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n   > \"Deliberate practice includes activities that have been specially designed to improve the current level of performance. These activities demand full concentration and effort and are not inherently enjoyable.\"\n\n2. Ericsson, K.A. & Pool, R. *Peak: Secrets from the New Science of Expertise*. Houghton Mifflin Harcourt, 2016.\n   > \"Mental representations are what distinguish experts from novices... The main thing that sets experts apart from the rest of us is that their years of practice have changed the neural circuitry in their brains to produce highly specialized mental representations, which in turn make possible the incredible memory, pattern recognition, problem solving, and other sorts of advanced abilities needed to excel in their particular specialties.\"\n\n**What is NOT cited and why:**\n\n- Gladwell, M. *Outliers* (2008) is deliberately excluded. Gladwell popularized the 10,000-hour figure while dropping the \"deliberate\" qualifier — the central mechanism of Ericsson's finding. Citing Gladwell would reproduce the distortion this skill is designed to correct.\n- Chase & Simon (1973) \"Perception in Chess\" (*Cognitive Psychology*, 4(1), 55–81) is the foundational precursor study on expert mental representations in chess — relevant but not cited directly because Ericsson's 1993 paper synthesizes and extends this work and is the canonical source for deliberate practice specifically.\n\nFile v1.0.2:examples/berlin-violin-study-1991-1993.md\n\n# Method in Action: Berlin Violin Study (1991–1993)\n\n> *Example for the [deliberate-practice](../SKILL.md) skill.*\n\n**Source:** Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C., \"The Role of Deliberate Practice in the Acquisition of Expert Performance,\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n\nEricsson and colleagues studied violin students at the Musikhochschule (Academy of Music) in West Berlin. Teachers nominated students into three groups based on their assessment of students' potential: (1) the \"best\" violinists — those judged capable of international solo careers; (2) \"good\" violinists — talented but below the top group; (3) music teachers — students training to teach, not perform at elite level.\n\nAll three groups had begun playing at approximately age 5. By age 20:\n\n- **Best violinists** had accumulated an estimated **10,000 hours** of deliberate practice (defined specifically as practice designed to improve performance, not performance itself or music-related activities like theory or group rehearsal).\n- **Good violinists** had accumulated approximately **8,000 hours**.\n- **Music teachers** had accumulated approximately **4,000 hours**.\n\nThe critical methodological point — erased by Malcolm Gladwell's 2008 popularization — is that Ericsson carefully distinguished *deliberate practice* (specifically designed, uncomfortable, feedback-rich, coach-structured) from total music-related activity time. The predictive variable was not \"hours of playing\" but \"hours of specifically deliberate practice.\" The best violinists also rated deliberate practice as significantly less enjoyable than performance, yet they allocated more time to it — evidence that intrinsic enjoyment does not drive deliberate practice; external structure and long-term goal commitment do.\n\nThis study is the primary empirical foundation for understanding expert performance. It shows that the representation gap between good and best performers is not mysterious — it is the cumulative product of structured, targeted, uncomfortable repetition designed to build increasingly sophisticated internal models.\n\nFile v1.0.2:skill-card.md\n\n## Description: <br>\nHelps an agent coach users through deliberate practice design when they face a skill plateau, need a high-performance learning program, or need to distinguish capability-building practice from routine execution. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nEmployees, external users, coaches, and developers use this skill to design targeted, feedback-rich practice routines for acquiring a skill instead of repeating comfortable work. It is most useful when a user can name an expert benchmark and wants a concrete practice design artifact. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: The skill can produce ineffective or misleading practice plans if the user lacks a real expert benchmark, reliable feedback source, or specific representation gap. <br>\nMitigation: Require the user to identify the expert standard, feedback latency, and target sub-skill before relying on the practice plan. <br>\nRisk: The skill asks reflective coaching questions and may surface sensitive performance or training details in the conversation. <br>\nMitigation: Avoid including private or regulated information unless it is necessary for the practice design and permitted by the deployment context. <br>\n\n\n## Reference(s): <br>\n- [Primary sources for deliberate practice](artifact/references/sources.md) <br>\n- [Berlin Violin Study example](artifact/examples/berlin-violin-study-1991-1993.md) <br>\n- [Ericsson, Krampe, and Tesch-Romer 1993](https://doi.org/10.1037/0033-295X.100.3.363) <br>\n- [ClawHub skill page](https://clawhub.ai/deciqai/skills/deliberate-practice) <br>\n- [Publisher profile](https://clawhub.ai/user/deciqai) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [text, markdown, guidance, configuration] <br>\n**Output Format:** [Markdown coaching prompts and a structured practice design artifact] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [May stop at explicit wait points while coaching a user step by step.] <br>\n\n## Skill Version(s): <br>\n1.0.2 (source: server release evidence) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>\n\nArchive v1.0.1: 5 files, 7589 bytes\n\nFiles: examples/berlin-violin-study-1991-1993.md (2172b), references/sources.md (1781b), skill-card.md (2310b), SKILL.md (7767b), _meta.json (138b)\n\nFile v1.0.1:SKILL.md\n\n---\nname: deliberate-practice\ndescription: \"Activate when: user says 'I've been doing this for years but I'm not getting better'; someone suspects a skill plateau despite continued effort; designing a learning program for a high-performance outcome; an organization reports high training hours but low skill transfer.\n  Do NOT activate when: goal is execution of existing skills rather than acquiring new ones (use deep-work instead); there is no identifiable expert performance benchmark to target.\"\n---\n\n# Deliberate Practice\n\n## Overview\n\nMost people confuse repetition with learning — they accumulate years of experience and plateau. Once an activity becomes automatic, executing it no longer builds new neural architecture. Ericsson, Krampe & Tesch-Römer (1993) showed the predictive variable is not hours of doing but hours of *specifically deliberate practice* — targeted, uncomfortable, feedback-rich repetition designed to build mental representations.\n\n**Cross-skill composition:** Use [`feedback-loops`](../feedback-loops/SKILL.md) first (audit your error signal); then [`metacognition`](../metacognition/SKILL.md) (surface your current representation gap); use instead of [`deep-work`](../deep-work/SKILL.md) when acquiring skills, not producing output; use alongside [`cognitive-evolution-stages`](../cognitive-evolution-stages/SKILL.md) for stage-aware practice design.\n\n---\n\n## When to Use\n\n**Trigger:** plateau despite experience; designing high-performance learning program; training hours high but skill transfer low; evaluating whether practice is building capability or maintaining it.\n**When NOT:** goal is execution not acquisition (use deep-work); no expert benchmark exists; bottleneck is motivational not representational.\n\n---\n\n## Coaching Novices (Adaptive Front Door)\n\n**Engine mode:** user has a concrete case → run The Process directly.\n**Coach mode:** user is unfamiliar or has no concrete case → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. Ask the plateau question: \"When you practice X, what does it feel like after 15 minutes — harder, the same, or easier?\" Comfort/easy = automatic = not building representations.\n2. Find the expert performance structure: \"Who is world-class at X? What do they perceive in the first 3 seconds that you don't?\" This locates the mental representation gap.\n3. Identify the discomfort zone: \"What part of practicing X makes you most want to stop?\" That is almost always where the gap lives.\n> **[WAIT — do not advance until user responds]**\n4. Design the smallest feedback loop: \"How would you know within 60 seconds whether a move was correct?\" Latency over 24h kills representation-building.\n> **[WAIT — do not advance until user responds]**\n5. Set the repetition target and stop-rule: \"How many reps of this specific discomfort can you sustain before concentration drops?\" (1–4 hours/day is Ericsson's ceiling.)\n> **[WAIT — do not advance until user responds]**\n\n---\n\n## The Process\n\n**Step 1 — Define the sub-skill with precision.** Not \"get better at X\" — specify the exact representational gap (e.g., \"detect when counterpart shifts from positional to interest-based\").\n**Step 2 — Find or construct the feedback mechanism.** Latency >24h breaks action-result association. Expert feedback > peer one level above > simulation with ground truth.\n**Step 3 — Diagnose the mental representation gap.** Ask: \"What does an expert *see* here that I don't?\" Not what they do — the doing follows from the seeing.\n**Step 4 — Design the repetition targeting the gap.** Must trigger the sub-skill, produce in-session feedback, and be executable at dozens–hundreds of reps per session.\n**Step 5 — Track representation progress, not output.** Output metrics lag by weeks. Track: \"Am I perceiving X earlier than before?\"\n**Step 6 — Apply the stop-rule.** Comfort = automaticity maintenance. Redesign to a harder sub-skill. End session when concentration drops.\n\n### Output: Practice Design Artifact\n\n```\nTarget sub-skill (precise): [specific representational gap]\nExpert mental representation: [what expert perceives that I currently don't]\nCurrent representation gap: [specific failure mode]\nFeedback — Source / Latency / Reliability:\nRepetition — Exercise / Volume / Duration / Frequency:\nProgress indicator (representation-level, not output): [what I will perceive by Week N]\nStop-rule triggers: comfortable → redesign; concentration drops → end; latency >24h → redesign\n```\n\n*→ Method in Action: [Berlin Violin Study (1991–1993)](examples/berlin-violin-study-1991-1993.md)*\n\n---\n\n## Practice Design Domain Packs\n\n**Medicine/Surgery:** sub-skill: laparoscopic tissue manipulation; feedback: simulator + debrief within 1h; rationalization to reject: \"I'll improve with more cases.\"\n**Writing:** sub-skill: eliminate nominalization in first-draft prose; feedback: rewrite published paragraphs vs original; rationalization: \"I write every day.\"\n**Investment:** sub-skill: identify customer concentration risk from footnotes in 20 min; feedback: 50-case retrospective library with outcomes.\n\nContribute packs via the deciqAI repo — requires sub-skill, expert representation, feedback latency, and common rationalization.\n\n---\n\n## Applying It Well\n\n- Target representations, not outcomes — ask \"What does the expert *perceive* that I don't?\"\n- Make feedback faster — redesign question: \"How do I get a reliable signal within 60 seconds?\"\n- Comfort signals time to redesign, not celebrate.\n- 1–4 genuine hours/day is Ericsson's hard ceiling; volume in degraded concentration reinforces errors.\n- Require expert think-alouds or annotated examples — you cannot design practice you cannot see.\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n---\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"I've been doing this for 10 years.\" | Duration is not deliberate practice. Years in automaticity = maintenance, not development. |\n| [D] \"I practice every day.\" | Comfortable daily repetition is automaticity reinforcement, not representation-building. |\n| [D] \"More cases/reps will help.\" | Only if structured to exceed current capability with rapid feedback. Otherwise more reps deepen the rut. |\n| [D] \"I can give myself feedback.\" | Self-feedback confirms what you already believe. External feedback from someone who sees the expert standard is required. |\n| [D] \"The discomfort means I'm doing it wrong.\" | Discomfort is the signal you are in deliberate practice. Comfort means automaticity. |\n| [D] \"My metrics are going up.\" | Output lags representation by weeks and is confounded by external factors. |\n| *→ Add [O] entries after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n---\n\n## Red Flags / Verification\n\n- Sessions feel comfortable — automaticity has absorbed the activity.\n- Feedback latency measured in days — action-result association cannot form.\n- Practitioner describes what expert *does* but not what they *perceive* — no representational target.\n- Practice volume cited as expertise without verifying hours were deliberate.\n- [ ] Sub-skill = specific representational gap; feedback latency <24h; expert representation identified.\n- [ ] Reps: dozens per session; stop-rule applied; progress tracked at representation level not output level.\n\n---\n\n*Part of **deciqAI Knowledge Skills** — open-source thinking skills that make rigor executable for AI agents. Built by deciqAI · https://deciqai.com · Contributions welcome — see the template at the repo root.*\n\nFile v1.0.1:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"deliberate-practice\",\n  \"version\": \"1.0.1\",\n  \"publishedAt\": 1783456318245\n}\n\nFile v1.0.1:references/sources.md\n\n# Sources — deliberate-practice\n\n> *Primary sources for the [deliberate-practice](../SKILL.md) skill.*\n\n**Primary sources with verbatim quotes:**\n\n1. Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C. \"The Role of Deliberate Practice in the Acquisition of Expert Performance.\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n   > \"Deliberate practice includes activities that have been specially designed to improve the current level of performance. These activities demand full concentration and effort and are not inherently enjoyable.\"\n\n2. Ericsson, K.A. & Pool, R. *Peak: Secrets from the New Science of Expertise*. Houghton Mifflin Harcourt, 2016.\n   > \"Mental representations are what distinguish experts from novices... The main thing that sets experts apart from the rest of us is that their years of practice have changed the neural circuitry in their brains to produce highly specialized mental representations, which in turn make possible the incredible memory, pattern recognition, problem solving, and other sorts of advanced abilities needed to excel in their particular specialties.\"\n\n**What is NOT cited and why:**\n\n- Gladwell, M. *Outliers* (2008) is deliberately excluded. Gladwell popularized the 10,000-hour figure while dropping the \"deliberate\" qualifier — the central mechanism of Ericsson's finding. Citing Gladwell would reproduce the distortion this skill is designed to correct.\n- Chase & Simon (1973) \"Perception in Chess\" (*Cognitive Psychology*, 4(1), 55–81) is the foundational precursor study on expert mental representations in chess — relevant but not cited directly because Ericsson's 1993 paper synthesizes and extends this work and is the canonical source for deliberate practice specifically.\n\nFile v1.0.1:examples/berlin-violin-study-1991-1993.md\n\n# Method in Action: Berlin Violin Study (1991–1993)\n\n> *Example for the [deliberate-practice](../SKILL.md) skill.*\n\n**Source:** Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C., \"The Role of Deliberate Practice in the Acquisition of Expert Performance,\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n\nEricsson and colleagues studied violin students at the Musikhochschule (Academy of Music) in West Berlin. Teachers nominated students into three groups based on their assessment of students' potential: (1) the \"best\" violinists — those judged capable of international solo careers; (2) \"good\" violinists — talented but below the top group; (3) music teachers — students training to teach, not perform at elite level.\n\nAll three groups had begun playing at approximately age 5. By age 20:\n\n- **Best violinists** had accumulated an estimated **10,000 hours** of deliberate practice (defined specifically as practice designed to improve performance, not performance itself or music-related activities like theory or group rehearsal).\n- **Good violinists** had accumulated approximately **8,000 hours**.\n- **Music teachers** had accumulated approximately **4,000 hours**.\n\nThe critical methodological point — erased by Malcolm Gladwell's 2008 popularization — is that Ericsson carefully distinguished *deliberate practice* (specifically designed, uncomfortable, feedback-rich, coach-structured) from total music-related activity time. The predictive variable was not \"hours of playing\" but \"hours of specifically deliberate practice.\" The best violinists also rated deliberate practice as significantly less enjoyable than performance, yet they allocated more time to it — evidence that intrinsic enjoyment does not drive deliberate practice; external structure and long-term goal commitment do.\n\nThis study is the primary empirical foundation for understanding expert performance. It shows that the representation gap between good and best performers is not mysterious — it is the cumulative product of structured, targeted, uncomfortable repetition designed to build increasingly sophisticated internal models.\n\nFile v1.0.1:skill-card.md\n\n## Description: <br>\nHelps an agent coach users through designing deliberate-practice routines for skill plateaus, using expert benchmarks, rapid feedback loops, targeted repetitions, and stop rules. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nEmployees, external users, coaches, trainers, and developers can use this skill to diagnose why repeated practice is not improving performance and to design a focused practice plan with fast feedback and representation-level progress checks. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Practice plans can be misleading when there is no clear expert benchmark or reliable feedback signal. <br>\nMitigation: Use the skill's verification checks to identify an expert representation, keep feedback latency under 24 hours, and review the resulting plan before relying on it. <br>\nRisk: The agent may over-continue a coaching flow that is intended to stop for user input. <br>\nMitigation: Respect the documented wait points and provide only the current question until the user responds. <br>\n\n\n## Reference(s): <br>\n- [Deliberate Practice release page](https://clawhub.ai/deciqai/skills/deliberate-practice) <br>\n- [Sources - deliberate-practice](references/sources.md) <br>\n- [Berlin Violin Study example](examples/berlin-violin-study-1991-1993.md) <br>\n- [Ericsson, Krampe, and Tesch-Romer 1993](https://doi.org/10.1037/0033-295X.100.3.363) <br>\n- [deciqAI](https://deciqai.com) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [text, markdown, guidance] <br>\n**Output Format:** [Markdown coaching prompts and a structured practice design artifact] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [May pause for user responses at marked coaching checkpoints before continuing.] <br>\n\n## Skill Version(s): <br>\n1.0.1 (source: server release evidence) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>\n\nArchive v1.0.0: 5 files, 7417 bytes\n\nFiles: examples/berlin-violin-study-1991-1993.md (2172b), references/sources.md (1781b), skill-card.md (1877b), SKILL.md (7767b), _meta.json (138b)\n\nFile v1.0.0:SKILL.md\n\n---\nname: deliberate-practice\ndescription: \"Activate when: user says 'I've been doing this for years but I'm not getting better'; someone suspects a skill plateau despite continued effort; designing a learning program for a high-performance outcome; an organization reports high training hours but low skill transfer.\n  Do NOT activate when: goal is execution of existing skills rather than acquiring new ones (use deep-work instead); there is no identifiable expert performance benchmark to target.\"\n---\n\n# Deliberate Practice\n\n## Overview\n\nMost people confuse repetition with learning — they accumulate years of experience and plateau. Once an activity becomes automatic, executing it no longer builds new neural architecture. Ericsson, Krampe & Tesch-Römer (1993) showed the predictive variable is not hours of doing but hours of *specifically deliberate practice* — targeted, uncomfortable, feedback-rich repetition designed to build mental representations.\n\n**Cross-skill composition:** Use [`feedback-loops`](../feedback-loops/SKILL.md) first (audit your error signal); then [`metacognition`](../metacognition/SKILL.md) (surface your current representation gap); use instead of [`deep-work`](../deep-work/SKILL.md) when acquiring skills, not producing output; use alongside [`cognitive-evolution-stages`](../cognitive-evolution-stages/SKILL.md) for stage-aware practice design.\n\n---\n\n## When to Use\n\n**Trigger:** plateau despite experience; designing high-performance learning program; training hours high but skill transfer low; evaluating whether practice is building capability or maintaining it.\n**When NOT:** goal is execution not acquisition (use deep-work); no expert benchmark exists; bottleneck is motivational not representational.\n\n---\n\n## Coaching Novices (Adaptive Front Door)\n\n**Engine mode:** user has a concrete case → run The Process directly.\n**Coach mode:** user is unfamiliar or has no concrete case → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. Ask the plateau question: \"When you practice X, what does it feel like after 15 minutes — harder, the same, or easier?\" Comfort/easy = automatic = not building representations.\n2. Find the expert performance structure: \"Who is world-class at X? What do they perceive in the first 3 seconds that you don't?\" This locates the mental representation gap.\n3. Identify the discomfort zone: \"What part of practicing X makes you most want to stop?\" That is almost always where the gap lives.\n> **[WAIT — do not advance until user responds]**\n4. Design the smallest feedback loop: \"How would you know within 60 seconds whether a move was correct?\" Latency over 24h kills representation-building.\n> **[WAIT — do not advance until user responds]**\n5. Set the repetition target and stop-rule: \"How many reps of this specific discomfort can you sustain before concentration drops?\" (1–4 hours/day is Ericsson's ceiling.)\n> **[WAIT — do not advance until user responds]**\n\n---\n\n## The Process\n\n**Step 1 — Define the sub-skill with precision.** Not \"get better at X\" — specify the exact representational gap (e.g., \"detect when counterpart shifts from positional to interest-based\").\n**Step 2 — Find or construct the feedback mechanism.** Latency >24h breaks action-result association. Expert feedback > peer one level above > simulation with ground truth.\n**Step 3 — Diagnose the mental representation gap.** Ask: \"What does an expert *see* here that I don't?\" Not what they do — the doing follows from the seeing.\n**Step 4 — Design the repetition targeting the gap.** Must trigger the sub-skill, produce in-session feedback, and be executable at dozens–hundreds of reps per session.\n**Step 5 — Track representation progress, not output.** Output metrics lag by weeks. Track: \"Am I perceiving X earlier than before?\"\n**Step 6 — Apply the stop-rule.** Comfort = automaticity maintenance. Redesign to a harder sub-skill. End session when concentration drops.\n\n### Output: Practice Design Artifact\n\n```\nTarget sub-skill (precise): [specific representational gap]\nExpert mental representation: [what expert perceives that I currently don't]\nCurrent representation gap: [specific failure mode]\nFeedback — Source / Latency / Reliability:\nRepetition — Exercise / Volume / Duration / Frequency:\nProgress indicator (representation-level, not output): [what I will perceive by Week N]\nStop-rule triggers: comfortable → redesign; concentration drops → end; latency >24h → redesign\n```\n\n*→ Method in Action: [Berlin Violin Study (1991–1993)](examples/berlin-violin-study-1991-1993.md)*\n\n---\n\n## Practice Design Domain Packs\n\n**Medicine/Surgery:** sub-skill: laparoscopic tissue manipulation; feedback: simulator + debrief within 1h; rationalization to reject: \"I'll improve with more cases.\"\n**Writing:** sub-skill: eliminate nominalization in first-draft prose; feedback: rewrite published paragraphs vs original; rationalization: \"I write every day.\"\n**Investment:** sub-skill: identify customer concentration risk from footnotes in 20 min; feedback: 50-case retrospective library with outcomes.\n\nContribute packs via the deciqAI repo — requires sub-skill, expert representation, feedback latency, and common rationalization.\n\n---\n\n## Applying It Well\n\n- Target representations, not outcomes — ask \"What does the expert *perceive* that I don't?\"\n- Make feedback faster — redesign question: \"How do I get a reliable signal within 60 seconds?\"\n- Comfort signals time to redesign, not celebrate.\n- 1–4 genuine hours/day is Ericsson's hard ceiling; volume in degraded concentration reinforces errors.\n- Require expert think-alouds or annotated examples — you cannot design practice you cannot see.\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n---\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"I've been doing this for 10 years.\" | Duration is not deliberate practice. Years in automaticity = maintenance, not development. |\n| [D] \"I practice every day.\" | Comfortable daily repetition is automaticity reinforcement, not representation-building. |\n| [D] \"More cases/reps will help.\" | Only if structured to exceed current capability with rapid feedback. Otherwise more reps deepen the rut. |\n| [D] \"I can give myself feedback.\" | Self-feedback confirms what you already believe. External feedback from someone who sees the expert standard is required. |\n| [D] \"The discomfort means I'm doing it wrong.\" | Discomfort is the signal you are in deliberate practice. Comfort means automaticity. |\n| [D] \"My metrics are going up.\" | Output lags representation by weeks and is confounded by external factors. |\n| *→ Add [O] entries after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n---\n\n## Red Flags / Verification\n\n- Sessions feel comfortable — automaticity has absorbed the activity.\n- Feedback latency measured in days — action-result association cannot form.\n- Practitioner describes what expert *does* but not what they *perceive* — no representational target.\n- Practice volume cited as expertise without verifying hours were deliberate.\n- [ ] Sub-skill = specific representational gap; feedback latency <24h; expert representation identified.\n- [ ] Reps: dozens per session; stop-rule applied; progress tracked at representation level not output level.\n\n---\n\n*Part of **deciqAI Knowledge Skills** — open-source thinking skills that make rigor executable for AI agents. Built by deciqAI · https://deciqai.com · Contributions welcome — see the template at the repo root.*\n\nFile v1.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"deliberate-practice\",\n  \"version\": \"1.0.0\",\n  \"publishedAt\": 1782544671993\n}\n\nFile v1.0.0:references/sources.md\n\n# Sources — deliberate-practice\n\n> *Primary sources for the [deliberate-practice](../SKILL.md) skill.*\n\n**Primary sources with verbatim quotes:**\n\n1. Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C. \"The Role of Deliberate Practice in the Acquisition of Expert Performance.\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n   > \"Deliberate practice includes activities that have been specially designed to improve the current level of performance. These activities demand full concentration and effort and are not inherently enjoyable.\"\n\n2. Ericsson, K.A. & Pool, R. *Peak: Secrets from the New Science of Expertise*. Houghton Mifflin Harcourt, 2016.\n   > \"Mental representations are what distinguish experts from novices... The main thing that sets experts apart from the rest of us is that their years of practice have changed the neural circuitry in their brains to produce highly specialized mental representations, which in turn make possible the incredible memory, pattern recognition, problem solving, and other sorts of advanced abilities needed to excel in their particular specialties.\"\n\n**What is NOT cited and why:**\n\n- Gladwell, M. *Outliers* (2008) is deliberately excluded. Gladwell popularized the 10,000-hour figure while dropping the \"deliberate\" qualifier — the central mechanism of Ericsson's finding. Citing Gladwell would reproduce the distortion this skill is designed to correct.\n- Chase & Simon (1973) \"Perception in Chess\" (*Cognitive Psychology*, 4(1), 55–81) is the foundational precursor study on expert mental representations in chess — relevant but not cited directly because Ericsson's 1993 paper synthesizes and extends this work and is the canonical source for deliberate practice specifically.\n\nFile v1.0.0:examples/berlin-violin-study-1991-1993.md\n\n# Method in Action: Berlin Violin Study (1991–1993)\n\n> *Example for the [deliberate-practice](../SKILL.md) skill.*\n\n**Source:** Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C., \"The Role of Deliberate Practice in the Acquisition of Expert Performance,\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n\nEricsson and colleagues studied violin students at the Musikhochschule (Academy of Music) in West Berlin. Teachers nominated students into three groups based on their assessment of students' potential: (1) the \"best\" violinists — those judged capable of international solo careers; (2) \"good\" violinists — talented but below the top group; (3) music teachers — students training to teach, not perform at elite level.\n\nAll three groups had begun playing at approximately age 5. By age 20:\n\n- **Best violinists** had accumulated an estimated **10,000 hours** of deliberate practice (defined specifically as practice designed to improve performance, not performance itself or music-related activities like theory or group rehearsal).\n- **Good violinists** had accumulated approximately **8,000 hours**.\n- **Music teachers** had accumulated approximately **4,000 hours**.\n\nThe critical methodological point — erased by Malcolm Gladwell's 2008 popularization — is that Ericsson carefully distinguished *deliberate practice* (specifically designed, uncomfortable, feedback-rich, coach-structured) from total music-related activity time. The predictive variable was not \"hours of playing\" but \"hours of specifically deliberate practice.\" The best violinists also rated deliberate practice as significantly less enjoyable than performance, yet they allocated more time to it — evidence that intrinsic enjoyment does not drive deliberate practice; external structure and long-term goal commitment do.\n\nThis study is the primary empirical foundation for understanding expert performance. It shows that the representation gap between good and best performers is not mysterious — it is the cumulative product of structured, targeted, uncomfortable repetition designed to build increasingly sophisticated internal models.\n\nFile v1.0.0:skill-card.md\n\n## Description: <br>\nHelps agents coach users through deliberate practice design for skill plateaus, expert-benchmark gaps, and feedback-rich training programs. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nEmployees, external users, and developers use this skill to turn vague skill-improvement goals into targeted practice plans with expert benchmarks, rapid feedback loops, repetition structure, and stop rules. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Review before execution as proposals could introduce incorrect or misleading guidance into skills. <br>\nMitigation: Review and scan skill before deployment. <br>\n\n## Reference(s): <br>\n- [Sources - deliberate-practice](references/sources.md) <br>\n- [Berlin Violin Study (1991-1993)](examples/berlin-violin-study-1991-1993.md) <br>\n- [The Role of Deliberate Practice in the Acquisition of Expert Performance](https://doi.org/10.1037/0033-295X.100.3.363) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [Text, Markdown, Guidance] <br>\n**Output Format:** [Markdown with structured coaching questions and a practice design artifact template] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [The skill may pause for user responses during step-by-step coaching and does not require files, credentials, accounts, or external tools.] <br>\n\n## Skill Version(s): <br>\n1.0.0 (source: server release evidence) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>","readmeExcerpt":"Skill: Deliberate Practice Owner: deciqai Summary: Activate when: user says 'I've been doing this for years but I'm not getting better'; someone suspects a skill plateau despite continued effort; designing a... Tags: latest:1.0.6 Version history: v1.0.6 | 2026-07-16T17:57:13.883Z | user Description tail link + agents machine-readable metadata line (deciqai.com/s/deliberate-practice.json) v1.0.5 | 2026-07-10T10:24:58.","codeSnippets":[],"executableExamples":[{"language":"text","snippet":"Target sub-skill (precise): [specific representational gap]\nExpert mental representation: [what expert perceives that I currently don't]\nCurrent representation gap: [specific failure mode]\nFeedback — Source / Latency / Reliability:\nRepetition — Exercise / Volume / Duration / Frequency:\nProgress indicator (representation-level, not output): [what I will perceive by Week N]\nStop-rule triggers: comfortable → redesign; concentration drops → end; latency >24h → redesign"},{"language":"text","snippet":"Target sub-skill (precise): [specific representational gap]\nExpert mental representation: [what expert perceives that I currently don't]\nCurrent representation gap: [specific failure mode]\nFeedback — Source / Latency / Reliability:\nRepetition — Exercise / Volume / Duration / Frequency:\nProgress indicator (representation-level, not output): [what I will perceive by Week N]\nStop-rule triggers: comfortable → redesign; concentration drops → end; latency >24h → redesign"},{"language":"text","snippet":"Target sub-skill (precise): [specific representational gap]\nExpert mental representation: [what expert perceives that I currently don't]\nCurrent representation gap: [specific failure mode]\nFeedback — Source / Latency / Reliability:\nRepetition — Exercise / Volume / Duration / Frequency:\nProgress indicator (representation-level, not output): [what I will perceive by Week N]\nStop-rule triggers: comfortable → redesign; concentration drops → end; latency >24h → redesign"},{"language":"text","snippet":"Target sub-skill (precise): [specific representational gap]\nExpert mental representation: [what expert perceives that I currently don't]\nCurrent representation gap: [specific failure mode]\nFeedback — Source / Latency / Reliability:\nRepetition — Exercise / Volume / Duration / Frequency:\nProgress indicator (representation-level, not output): [what I will perceive by Week N]\nStop-rule triggers: comfortable → redesign; concentration drops → end; latency >24h → redesign"},{"language":"text","snippet":"Target sub-skill (precise): [specific representational gap]\nExpert mental representation: [what expert perceives that I currently don't]\nCurrent representation gap: [specific failure mode]\nFeedback — Source / Latency / Reliability:\nRepetition — Exercise / Volume / Duration / Frequency:\nProgress indicator (representation-level, not output): [what I will perceive by Week N]\nStop-rule triggers: comfortable → redesign; concentration drops → end; latency >24h → redesign"},{"language":"text","snippet":"Target sub-skill (precise): [specific representational gap]\nExpert mental representation: [what expert perceives that I currently don't]\nCurrent representation gap: [specific failure mode]\nFeedback — Source / Latency / Reliability:\nRepetition — Exercise / Volume / Duration / Frequency:\nProgress indicator (representation-level, not output): [what I will perceive by Week N]\nStop-rule triggers: comfortable → redesign; concentration drops → end; latency >24h → redesign"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: deliberate-practice\ndescription: \"Activate when: user says 'I've been doing this for years but I'm not getting better'; someone suspects a skill plateau despite continued effort; designing a learning program for a high-performance outcome; an organization reports high training hours but low skill transfer.\n  Do NOT activate when: goal is execution of existing skills rather than acquiring new ones (use deep-work instead); there is no identifiable expert performance benchmark to target. More: deciqai.com/c/deliberate-practice\"\n---\n\n# Deliberate Practice\n\n## Overview\n\nMost people confuse repetition with learning — they accumulate years of experience and plateau. Once an activity becomes automatic, executing it no longer builds new neural architecture. Ericsson, Krampe & Tesch-Römer (1993) showed the predictive variable is not hours of doing but hours of *specifically deliberate practice* — targeted, uncomfortable, feedback-rich repetition designed to build mental representations.\n\n**Cross-skill composition:** Use `feedback-loops` first (audit your error signal); then `metacognition` (surface your current representation gap); use instead of `deep-work` when acquiring skills, not producing output; use alongside `cognitive-evolution-stages` for stage-aware practice design.\n\n---\n\n## When to Use\n\n**Trigger:** plateau despite experience; designing high-performance learning program; training hours high but skill transfer low; evaluating whether practice is building capability or maintaining it; skill atrophy or deskilling as AI copilots absorb the routine reps (AI adoption, AI hype, \"will AI make me worse at my craft\").\n**When NOT:** goal is execution not acquisition (use deep-work); no expert benchmark exists; bottleneck is motivational not representational.\n\n---\n\n## Coaching Novices (Adaptive Front Door)\n\n**Engine mode:** user has a concrete case → run The Process directly.\n**Coach mode:** user is unfamiliar or has no concrete case → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. Ask the plateau question: \"When you practice X, what does it feel like after 15 minutes — harder, the same, or easier?\" Comfort/easy = automatic = not building representations.\n2. Find the expert performance structure: \"Who is world-class at X? What do they perceive in the first 3 seconds that you don't?\" This locates the mental representation gap.\n3. Identify the discomfort zone: \"What part of practicing X makes you most want to stop?\" That is almost always where the gap lives.\n> **[WAIT — do not advance until user responds]**\n4. Design the smallest feedback loop: \"How would you know within 60 seconds whether a move was correct?\" Latency over 24h kills representation-building.\n> **[WAIT — do not advance until user responds]**\n5. Set the repetition target and stop-rule: \"How many reps of this specific discomfort can you sustain before concentration drops?\" (1–4 hours/day is E"},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"deliberate-practice\",\n  \"version\": \"1.0.6\",\n  \"publishedAt\": 1784224633883\n}"},{"path":"references/sources.md","content":"# Sources — deliberate-practice\n\n> *Primary sources for the [deliberate-practice](../SKILL.md) skill.*\n\n**Primary sources with verbatim quotes:**\n\n1. Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C. \"The Role of Deliberate Practice in the Acquisition of Expert Performance.\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n   > \"Deliberate practice includes activities that have been specially designed to improve the current level of performance. These activities demand full concentration and effort and are not inherently enjoyable.\"\n\n2. Ericsson, K.A. & Pool, R. *Peak: Secrets from the New Science of Expertise*. Houghton Mifflin Harcourt, 2016.\n   > \"Mental representations are what distinguish experts from novices... The main thing that sets experts apart from the rest of us is that their years of practice have changed the neural circuitry in their brains to produce highly specialized mental representations, which in turn make possible the incredible memory, pattern recognition, problem solving, and other sorts of advanced abilities needed to excel in their particular specialties.\"\n\n3. Franklin, B. *The Autobiography of Benjamin Franklin*. Part One, written 1771. Widely available; e.g. Project Gutenberg ebook #148.\n   > \"Then I compared my Spectator with the original, discovered some of my faults, and corrected them.\"\n\n**Contemporary context sources (2024–2026 AI-adoption example):**\n\n4. Stack Overflow. *2024 Developer Survey* (AI section). Reported that a large majority of professional developers were using or planning to use AI tools in their development workflow. https://survey.stackoverflow.co/2024/\n   - Used only to establish the durable, widely-reported fact that AI coding/writing assistants became mainstream in professional workflows by 2024–2025. No precise percentages are asserted in the example beyond \"a large majority.\"\n\n5. GitHub. \"Research: Quantifying GitHub Copilot's impact on developer productivity and happiness.\" GitHub Blog, 2022. https://github.blog/2022-09-07-research-quantifying-github-copilots-impact-on-developer-productivity-and-happiness/\n   - Cited as background evidence for Copilot's productivity effect and its adoption as a routine tool; the deskilling/atrophy risk in the example is a *prediction from Ericsson's model*, not a claim of a measured 2024–2026 effect.\n\n**What is NOT cited and why:**\n\n- Gladwell, M. *Outliers* (2008) is deliberately excluded. Gladwell popularized the 10,000-hour figure while dropping the \"deliberate\" qualifier — the central mechanism of Ericsson's finding. Citing Gladwell would reproduce the distortion this skill is designed to correct.\n- Chase & Simon (1973) \"Perception in Chess\" (*Cognitive Psychology*, 4(1), 55–81) is the foundational precursor study on expert mental representations in chess — relevant but not cited directly because Ericsson's 1993 paper synthesizes and extends this work and is the canonical source for deliberate practice specifically"},{"path":"examples/berlin-violin-study-1991-1993.md","content":"# Method in Action: Berlin Violin Study (1991–1993)\n\n> *Example for the [deliberate-practice](../SKILL.md) skill.*\n\n**Source:** Ericsson, K.A., Krampe, R.T. & Tesch-Römer, C., \"The Role of Deliberate Practice in the Acquisition of Expert Performance,\" *Psychological Review*, 100(3), 363–406, 1993. https://doi.org/10.1037/0033-295X.100.3.363\n\nEricsson and colleagues studied violin students at the Musikhochschule (Academy of Music) in West Berlin. Teachers nominated students into three groups based on their assessment of students' potential: (1) the \"best\" violinists — those judged capable of international solo careers; (2) \"good\" violinists — talented but below the top group; (3) music teachers — students training to teach, not perform at elite level.\n\nAll three groups had begun playing at approximately age 5. By age 20:\n\n- **Best violinists** had accumulated an estimated **10,000 hours** of deliberate practice (defined specifically as practice designed to improve performance, not performance itself or music-related activities like theory or group rehearsal).\n- **Good violinists** had accumulated approximately **8,000 hours**.\n- **Music teachers** had accumulated approximately **4,000 hours**.\n\nThe critical methodological point — erased by Malcolm Gladwell's 2008 popularization — is that Ericsson carefully distinguished *deliberate practice* (specifically designed, uncomfortable, feedback-rich, coach-structured) from total music-related activity time. The predictive variable was not \"hours of playing\" but \"hours of specifically deliberate practice.\" The best violinists also rated deliberate practice as significantly less enjoyable than performance, yet they allocated more time to it — evidence that intrinsic enjoyment does not drive deliberate practice; external structure and long-term goal commitment do.\n\nThis study is the primary empirical foundation for understanding expert performance. It shows that the representation gap between good and best performers is not mysterious — it is the cumulative product of structured, targeted, uncomfortable repetition designed to build increasingly sophisticated internal models."},{"path":"examples/franklin-spectator-writing-method.md","content":"# Method in Action: Benjamin Franklin's Spectator Writing Method (c. 1718–1723)\n\n> *Example for the [deliberate-practice](../SKILL.md) skill.*\n\nAs a teenage printer's apprentice in Boston, Benjamin Franklin engineered — two centuries before Ericsson named the mechanism — a complete deliberate practice system for prose writing, documented in detail in his *Autobiography*. It is the canonical historical case of self-designed practice in a domain with no available coach.\n\n**Step 1 — Define the sub-skill with precision.** Franklin did not resolve to \"write better.\" His father had compared his letters (from a written debate with his friend John Collins) against Collins's and identified specific deficits: Franklin's writing fell short in elegance of expression, in method, and in clarity. Franklin took these as three distinct representational gaps — word choice, arrangement of thoughts, and expression — and attacked each separately.\n\n**Step 2 — Find or construct the feedback mechanism.** With no writing tutor available, Franklin constructed a ground-truth comparator: an odd volume of *The Spectator*, the London periodical of Addison and Steele, whose prose he judged excellent. The original essay itself became the expert benchmark against which every attempt could be scored — feedback latency of minutes, not days.\n\n**Step 3 — Diagnose the mental representation gap.** The exercise structure forced the diagnosis. Franklin made short hints of the sentiment of each sentence in an essay, set them aside for a few days, then attempted to reconstruct the full essay from the hints in his own words. Comparing his reconstruction against Addison's original exposed exactly where his internal model of good prose diverged from the expert's — fault by fault, sentence by sentence.\n\n**Step 4 — Design the repetition targeting the gap.** When comparison revealed his stock of words was too small, Franklin redesigned the drill: he turned Spectator essays into verse and, after forgetting the originals, back into prose — because versification forces a continual search for words of different lengths and sounds to fit meter and rhyme. When the gap was arrangement rather than vocabulary, he jumbled his sentence hints into confusion, waited weeks, and then attempted to reorder them into the best structure before reconstructing — a drill isolating the method-of-thought sub-skill alone.\n\n**Step 5 — Track representation progress, not output.** Franklin's progress measure was perceptual, not productive: he tracked the faults he could now *discover* in his own reconstructions and correct. He records that he sometimes had the pleasure of judging that, in small particulars, he had improved on the original's method or language — evidence his internal representation of good prose had begun to match, and locally exceed, the benchmark.\n\n**Step 6 — Apply the stop-rule.** The escalation from plain reconstruction, to verse conversion, to jumbled-hint reordering is the stop-rule in action: each t"}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":"Activate when: user says 'I've been doing this for years but I'm not getting better'; someone suspects a skill plateau despite continued effort; designing a... Skill: Deliberate Practice Owner: deciqai Summary: Activate when: user says 'I've been doing this for years but I'm not getting better'; someone suspects a skill plateau despite continued effort; designing a... Tags: latest:1.0.6 Version history: v1.0.6 | 2026-07-16T17:57:13.883Z | user Description tail link + agents machine-readable metadata line (deciqai.com/s/deliberate-practice.json) v1.0.5 | 2026-07-10T10:24:58.","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":2004,"uniquenessScore":50,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-11T06:20:09.997Z","emptyReason":"No screenshots, media assets, or demo links are available."},"primaryImageUrl":null,"mediaAssetCount":0,"assets":[],"demoUrl":null},"ownerResources":{"evidence":{"source":"unclaimed","verified":false,"confidence":"low","updatedAt":"2026-10-11T06:20:09.997Z","emptyReason":"This page has not been claimed by the agent owner."},"hasCustomPage":false,"customPageUpdatedAt":null,"customLinks":[],"structuredLinks":{"docsUrl":null,"demoUrl":null,"supportUrl":null,"pricingUrl":null,"statusUrl":null},"customPage":null},"relatedAgents":{"evidence":{"source":"protocol-neighbors","verified":false,"confidence":"medium","updatedAt":"2026-10-11T08:41:39.360Z","emptyReason":null},"items":[{"id":"8ebccd8e-3863-4187-8355-c3f14e1f9edf","entityType":"agent","canonicalPath":"/agent/iofficeai-aionui","slug":"iofficeai-aionui","name":"AionUi","description":"Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!","url":"https://github.com/iOfficeAI/AionUi","homepage":"https://www.aionui.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-10-09T19:11:12.944Z","createdAt":"2026-02-25T03:38:16.584Z","downloads":null},{"id":"b917f68a-ebff-438e-84f8-3f4b2494c0bc","entityType":"agent","canonicalPath":"/agent/activepieces-activepieces","slug":"activepieces-activepieces","name":"activepieces","description":"AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents","url":"https://github.com/activepieces/activepieces","homepage":"https://www.activepieces.com","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-15T02:22:12.426Z","createdAt":"2026-02-25T03:38:12.412Z","downloads":null},{"id":"5cb26759-3a39-483f-94cf-276a98c13bb8","entityType":"agent","canonicalPath":"/agent/cherryhq-cherry-studio","slug":"cherryhq-cherry-studio","name":"cherry-studio","description":"AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs","url":"https://github.com/CherryHQ/cherry-studio","homepage":"https://cherry-ai.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-11T14:38:40.986Z","createdAt":"2026-02-25T03:38:19.379Z","downloads":null},{"id":"6f6582d0-5d76-4f0f-b81d-86520247950b","entityType":"agent","canonicalPath":"/agent/copilotkit-copilotkit","slug":"copilotkit-copilotkit","name":"CopilotKit","description":"The Frontend for Agents & Generative UI. React + Angular","url":"https://github.com/CopilotKit/CopilotKit","homepage":"https://docs.copilotkit.ai","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-03-25T09:50:57.846Z","createdAt":"2026-02-25T03:39:14.617Z","downloads":null}],"links":{"hub":"/agent","source":"/agent/source/clawhub","protocols":[{"label":"OpenClaw","href":"/agent/protocol/openclew"}]}}}