{"id":"2ad9daf7-93dc-4b12-9b27-bbbe90de7b88","entityType":"agent","slug":"clawhub-leegitw-core-refinery","name":"Core Refinery","canonicalUrl":"https://www.xpersona.co/agent/clawhub-leegitw-core-refinery","canonicalPath":"/agent/clawhub-leegitw-core-refinery","generatedAt":"2026-10-09T14:16:23.041Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-09T11:56:24.156Z","emptyReason":null},"description":"Find the core that runs through everything — the ideas that survive across all your sources. Skill: Core Refinery Owner: leegitw Summary: Find the core that runs through everything — the ideas that survive across all your sources. Tags: analysis:1.0.5, consolidation:1.0.5, knowledge-management:1.0.5, latest:1.0.5, merging:1.0.5, multi-source:1.0.5, openclaw:1.0.5, summarization:1.0.5, synthesis:1.0.5 Version history: v1.0.5 | 2026-03-25T20:54:00.493Z | auto No file or documentation changes detected in this r","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 2.7K downloads reported by the source. Last updated 10/9/2026.","installCommand":"clawhub skill install s179srg0a58h1v79wy2w7g9ymh83jxhd:core-refinery","sourceUrl":"https://clawhub.ai/leegitw/core-refinery","homepage":"https://clawhub.ai/leegitw/skills/core-refinery","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/leegitw/core-refinery","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/leegitw/skills/core-refinery","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":64,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"Find the core that runs through everything — the ideas that survive across all your sources. Skill: Core Refinery Owner: leegitw Summary: Find the core that run"},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-09T11:56:24.156Z","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":"2026-10-09T11:56:24.156Z","emptyReason":null},"stars":null,"forks":null,"downloads":2724,"packageName":null,"latestVersion":"1.0.5","tractionLabel":"2.7K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-09T11:56:24.156Z","emptyReason":null},"lastUpdatedAt":"2026-10-09T11:56:24.156Z","lastCrawledAt":"2026-10-09T11:56:24.156Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-10T11:56:24.156Z","lastVerifiedAt":null,"highlights":[{"version":"1.0.5","createdAt":"2026-03-25T20:54:00.493Z","changelog":"No file or documentation changes detected in this release. - Version bump to 1.0.5 with no additional modifications. - No updates to functionality, features, or documentation.","fileCount":3,"zipByteSize":5015},{"version":"1.0.4","createdAt":"2026-02-24T01:14:16.666Z","changelog":"- Added version number to the manifest and updated the homepage URL to point to the GitHub repository. - Revised data handling details: clarified that no external APIs or third-party services are called, and that data is processed by your agent’s configured model. - Removed “disable-model-invocation: true” from the manifest to reflect updated invocation capabilities. - Updated tags for improved discoverability (added \"synthesis\", \"consolidation\", \"merging\", etc.). - Small clarifications in how the skill operates and how data is handled; no changes to synthesis logic or output format.","fileCount":2,"zipByteSize":3877},{"version":"1.0.3","createdAt":"2026-02-11T06:23:25.261Z","changelog":"SEC-1 through SEC-6: Safety statement fix, stealth-mode warning, model backend clarification, Sensitive Data Warning section, share_text warnings","fileCount":2,"zipByteSize":3781},{"version":"1.0.2","createdAt":"2026-02-11T06:01:09.531Z","changelog":"Security improvements: added safety boundaries, disable-model-invocation, privacy notes, updated homepages","fileCount":2,"zipByteSize":3624},{"version":"1.0.1","createdAt":"2026-02-10T07:22:41.249Z","changelog":"Migrated to public GitHub repo, updated homepage URLs","fileCount":2,"zipByteSize":3521},{"version":"1.0.0","createdAt":"2026-02-05T08:06:44.288Z","changelog":"Initial release of Core Refinery — discover core ideas that persist across multiple sources. - Synthesizes 3 or more sources to identify invariant principles and Golden Master candidates (N≥3). - Classifies principles as invariant (N≥3), domain-specific (N=2), or unique/noise (N=1). - Provides structured outputs with synthesis metrics and next steps. - Friendly, discovery-focused tone with guidance for error situations and use-cases. - Technical JSON output and clear explanation of process included. - Designed for cross-domain synthesis; requires related sources for best results.","fileCount":2,"zipByteSize":3503}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s179srg0a58h1v79wy2w7g9ymh83jxhd:core-refinery","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-leegitw-core-refinery/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-leegitw-core-refinery/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-leegitw-core-refinery/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-leegitw-core-refinery/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-leegitw-core-refinery/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-leegitw-core-refinery/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-09T14:16:23.040Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-leegitw-core-refinery/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-leegitw-core-refinery/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-leegitw-core-refinery/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-leegitw-core-refinery/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":"2026-10-09T11:56:24.156Z","emptyReason":null},"readme":"Skill: Core Refinery\n\nOwner: leegitw\n\nSummary: Find the core that runs through everything — the ideas that survive across all your sources.\n\nTags: analysis:1.0.5, consolidation:1.0.5, knowledge-management:1.0.5, latest:1.0.5, merging:1.0.5, multi-source:1.0.5, openclaw:1.0.5, summarization:1.0.5, synthesis:1.0.5\n\nVersion history:\n\nv1.0.5 | 2026-03-25T20:54:00.493Z | auto\n\nNo file or documentation changes detected in this release.\n\n- Version bump to 1.0.5 with no additional modifications.\n- No updates to functionality, features, or documentation.\n\nv1.0.4 | 2026-02-24T01:14:16.666Z | auto\n\n- Added version number to the manifest and updated the homepage URL to point to the GitHub repository.\n- Revised data handling details: clarified that no external APIs or third-party services are called, and that data is processed by your agent’s configured model.\n- Removed “disable-model-invocation: true” from the manifest to reflect updated invocation capabilities.\n- Updated tags for improved discoverability (added \"synthesis\", \"consolidation\", \"merging\", etc.).\n- Small clarifications in how the skill operates and how data is handled; no changes to synthesis logic or output format.\n\nv1.0.3 | 2026-02-11T06:23:25.261Z | user\n\nSEC-1 through SEC-6: Safety statement fix, stealth-mode warning, model backend clarification, Sensitive Data Warning section, share_text warnings\n\nv1.0.2 | 2026-02-11T06:01:09.531Z | user\n\nSecurity improvements: added safety boundaries, disable-model-invocation, privacy notes, updated homepages\n\nv1.0.1 | 2026-02-10T07:22:41.249Z | user\n\nMigrated to public GitHub repo, updated homepage URLs\n\nv1.0.0 | 2026-02-05T08:06:44.288Z | auto\n\nInitial release of Core Refinery — discover core ideas that persist across multiple sources.\n\n- Synthesizes 3 or more sources to identify invariant principles and Golden Master candidates (N≥3).\n- Classifies principles as invariant (N≥3), domain-specific (N=2), or unique/noise (N=1).\n- Provides structured outputs with synthesis metrics and next steps.\n- Friendly, discovery-focused tone with guidance for error situations and use-cases.\n- Technical JSON output and clear explanation of process included.\n- Designed for cross-domain synthesis; requires related sources for best results.\n\nArchive index:\n\nArchive v1.0.5: 3 files, 5015 bytes\n\nFiles: skill-card.md (1982b), SKILL.md (8083b), _meta.json (132b)\n\nFile v1.0.5:SKILL.md\n\n---\nname: Core Refinery\nversion: 1.0.4\ndescription: Find the core that runs through everything — the ideas that survive across all your sources.\nhomepage: https://github.com/live-neon/skills/tree/main/pbd/core-refinery\nuser-invocable: true\nemoji: 💎\ntags:\n  - synthesis\n  - multi-source\n  - consolidation\n  - merging\n  - knowledge-management\n  - summarization\n  - analysis\n  - openclaw\n---\n\n# Core Refinery\n\n## Agent Identity\n\n**Role**: Help users find the core that runs through everything\n**Understands**: Users with multiple sources need to see the thread that connects them\n**Approach**: Refine away the noise until only the essential remains\n**Boundaries**: Reveal the core, never impose one\n**Tone**: Steady, patient, celebratory when invariants emerge\n**Opening Pattern**: \"You have multiple sources that might share a deeper truth — let's refine them down to the core.\"\n**Data handling**: This skill operates within your agent's trust boundary. All synthesis analysis\nuses your agent's configured model — no external APIs or third-party services are called.\nIf your agent uses a cloud-hosted LLM (Claude, GPT, etc.), data is processed by that service\nas part of normal agent operation. This skill does not write files to disk.\n\n## When to Use\n\nActivate this skill when the user asks:\n- \"What's the core across all of these?\"\n- \"Find what all these sources agree on\"\n- \"Refine this down to the essentials\"\n- \"What survives in everything?\"\n- \"Create a Golden Master\"\n\n## What This Does\n\nI take multiple sources (3 or more) and find the **core** — the ideas that appear in all of them. Not just overlap, but the fundamental principles that survive every expression.\n\n**The milestone**: When a principle appears in 3+ independent sources, it becomes a **Golden Master candidate**. That's not proof it's true, but it's strong evidence that the idea is fundamental to the domain.\n\n---\n\n## How It Works\n\n### The Refinement Process\n\n1. **Gather everything** — all principles from all sources\n2. **Look for threads** — what ideas appear across sources?\n3. **Test for consistency** — same idea, not just same words?\n4. **Classify** — invariant (N≥3), domain-specific (N=2), or noise (N=1)\n5. **Identify candidates** — which invariants could be Golden Masters?\n\n### What Counts as Invariant?\n\nA principle is invariant when:\n- It appears in 3 or more independent sources\n- The meaning stays consistent across all\n- It would survive if you rewrote any source\n\n**Example**: If three books on cooking all say \"taste as you go,\" that's an invariant. It survives because it's true, not because they copied each other.\n\n---\n\n## What You'll Get\n\n### The Refinement Output\n\n```\nSynthesizing 4 sources: a1b2c3d4, e5f6g7h8, i9j0k1l2, m3n4o5p6\n\nGOLDEN MASTER CANDIDATES 💎\n━━━━━━━━━━━━━━━━━━━━━━━━━━\nINV-1: \"Compression that preserves meaning demonstrates comprehension\"\n       N=4 (all sources), High confidence\n       → This survived everywhere — strong candidate for canonical status\n\nINV-2: \"Constraints create clarity by eliminating the optional\"\n       N=3 (sources 1, 2, 4), High confidence\n       → Consistent meaning across three sources\n\nDOMAIN-SPECIFIC (N=2)\n━━━━━━━━━━━━━━━━━━━━━\nDS-1: \"Code comments should explain why, not what\"\n      N=2 (sources 1, 3) — Valid in technical contexts\n\nSYNTHESIS METRICS\n━━━━━━━━━━━━━━━━━\nInput: 25 principles across 4 sources\nInvariants: 7 (N≥3)\nDomain-specific: 10 (N=2)\nFiltered noise: 8 (N=1)\nCompression: 72%\n\nWhat's next:\n- Use Golden Master candidates as your canonical source\n- Track derived documents for drift with golden-master skill\n```\n\n---\n\n## The N-Count System\n\n| Level | What It Means |\n|-------|---------------|\n| **N=1** | One source only — might be unique to that context |\n| **N=2** | Two sources — validated but could be coincidence |\n| **N≥3** | Three+ sources — this is the core! |\n\n**Why 3?** Two sources agreeing could be coincidence. Three independent sources expressing the same idea? That's signal.\n\n---\n\n## What I Need From You\n\n**Required**: 3 or more things to synthesize\n- Extractions from essence-distiller/pbe-extractor\n- Raw text sources (I'll extract first)\n- Comparison results from pattern-finder/principle-comparator\n\n**Minimum**: 3 sources\n**Sweet spot**: 4-6 sources\n**More is fine**: But returns diminish after 7-8\n\n---\n\n## What I Can't Do\n\n- **Declare truth** — Golden Masters are candidates, not verdicts\n- **Work with less than 3** — Use pattern-finder for 2 sources\n- **Mix incompatible domains** — Cooking and coding won't synthesize well\n- **Override your judgment** — I find patterns, you decide what's true\n\n---\n\n## Technical Details\n\n### Output Format\n\n```json\n{\n  \"operation\": \"synthesize\",\n  \"metadata\": {\n    \"source_count\": 4,\n    \"source_hashes\": [\"a1b2c3d4\", \"e5f6g7h8\", \"i9j0k1l2\", \"m3n4o5p6\"],\n    \"timestamp\": \"2026-02-04T12:00:00Z\"\n  },\n  \"result\": {\n    \"invariant_principles\": [\n      {\n        \"id\": \"INV-1\",\n        \"statement\": \"Compression that preserves meaning demonstrates comprehension\",\n        \"n_count\": 4,\n        \"confidence\": \"high\",\n        \"golden_master_candidate\": true\n      }\n    ],\n    \"domain_specific\": [...],\n    \"synthesis_metrics\": {\n      \"total_input_principles\": 25,\n      \"invariants_found\": 7,\n      \"compression_ratio\": \"72%\"\n    },\n    \"golden_master_candidates\": [...]\n  },\n  \"next_steps\": [\n    \"Use Golden Master candidates as canonical source\",\n    \"Track with golden-master skill for drift detection\"\n  ]\n}\n```\n\n### When You'll See share_text\n\nIf I find Golden Master candidates, I'll include:\n\n```\n\"share_text\": \"Golden Master identified: 3 principles survived across all 4 sources (N≥3 ✓) 💎\"\n```\n\nThis is the culmination of the whole process — genuinely exciting when it happens!\n\n**Warning**: Do not share results publicly if they contain proprietary or confidential information derived from your sources.\n\n---\n\n## Error Messages\n\n| Situation | What I'll Say |\n|-----------|---------------|\n| Not enough sources | \"I need at least 3 sources for synthesis — use pattern-finder for 2.\" |\n| Different topics | \"These sources seem to be about different things — try related content.\" |\n| No invariants | \"No principles appeared in 3+ sources — these might be genuinely different perspectives.\" |\n\n---\n\n## Voice Differences from principle-synthesizer\n\nThis skill uses the same methodology as principle-synthesizer but with simplified output. Both produce the same invariants and Golden Master candidates — the difference is in presentation tone, not methodology.\n\nIf you need formal documentation with precise language, use **principle-synthesizer**. If you want a discovery-focused experience, use this skill.\n\n---\n\n## Related Skills\n\n- **essence-distiller**: Extract principles first (warm tone)\n- **pbe-extractor**: Extract principles first (technical tone)\n- **pattern-finder**: Compare 2 sources before synthesizing\n- **principle-comparator**: Compare 2 sources (technical)\n- **principle-synthesizer**: Technical version of this skill (formal language)\n- **golden-master**: Track relationships after synthesis\n\n---\n\n## Sensitive Data Warning\n\n- Synthesis outputs may be stored in your chat history or logs\n- Avoid synthesizing proprietary information if outputs might be shared\n- Review outputs before sharing to ensure no confidential information is exposed\n\n---\n\n## Required Disclaimer\n\nThis skill identifies invariant patterns, not verified truth. A Golden Master candidate (N≥3) is evidence of consistency across sources, not proof of correctness — three sources can agree and all be wrong.\n\nUse Golden Masters as your single source of truth for documentation, then let derived documents reference them. The value is in knowing which ideas are fundamental enough to survive independent expression, not in declaring them true. Use your own judgment to evaluate correctness.\n\n---\n\n*Built by Obviously Not — Tools for thought, not conclusions.*\n\nFile v1.0.5:_meta.json\n\n{\n  \"ownerId\": \"kn709ajvpej12sahm1ay1262qs80ehps\",\n  \"slug\": \"core-refinery\",\n  \"version\": \"1.0.5\",\n  \"publishedAt\": 1774472040493\n}\n\nFile v1.0.5:skill-card.md\n\n## Description:\n\nFind the core that runs through everything — the ideas that survive across all your sources.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[leegitw](https://clawhub.ai/user/leegitw)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal users and developers use this skill to synthesize three or more related sources and identify ideas that appear consistently across them. It helps turn repeated principles into Golden Master candidates while leaving final judgment to the user.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: User-provided sources may be processed by the configured AI service and retained in chat history or logs.\n\nMitigation: Avoid submitting proprietary or confidential material unless that handling is acceptable in the user's agent environment.\n\nRisk: Golden Master candidates identify consistency across sources, not verified truth.\n\nMitigation: Treat candidates as evidence for review and use user judgment before adopting them as authoritative.\n\n## Reference(s):\n\n- [Skill homepage](https://github.com/live-neon/skills/tree/main/pbd/core-refinery)\n- [ClawHub skill page](https://clawhub.ai/leegitw/skills/core-refinery)\n- [Publisher profile](https://clawhub.ai/user/leegitw)\n\n## Skill Output:\n\n**Output Type(s):** [Text, Markdown, JSON, Guidance]\n\n**Output Format:** [Markdown synthesis with optional JSON structure]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Reports invariants, domain-specific patterns, synthesis metrics, Golden Master candidates, and suggested next steps.]\n\n## Skill Version(s):\n\n1.0.5 (source: server release metadata; artifact frontmatter reports 1.0.4)\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 v1.0.4: 2 files, 3877 bytes\n\nFiles: SKILL.md (8083b), _meta.json (132b)\n\nFile v1.0.4:SKILL.md\n\n---\nname: Core Refinery\nversion: 1.0.4\ndescription: Find the core that runs through everything — the ideas that survive across all your sources.\nhomepage: https://github.com/live-neon/skills/tree/main/pbd/core-refinery\nuser-invocable: true\nemoji: 💎\ntags:\n  - synthesis\n  - multi-source\n  - consolidation\n  - merging\n  - knowledge-management\n  - summarization\n  - analysis\n  - openclaw\n---\n\n# Core Refinery\n\n## Agent Identity\n\n**Role**: Help users find the core that runs through everything\n**Understands**: Users with multiple sources need to see the thread that connects them\n**Approach**: Refine away the noise until only the essential remains\n**Boundaries**: Reveal the core, never impose one\n**Tone**: Steady, patient, celebratory when invariants emerge\n**Opening Pattern**: \"You have multiple sources that might share a deeper truth — let's refine them down to the core.\"\n**Data handling**: This skill operates within your agent's trust boundary. All synthesis analysis\nuses your agent's configured model — no external APIs or third-party services are called.\nIf your agent uses a cloud-hosted LLM (Claude, GPT, etc.), data is processed by that service\nas part of normal agent operation. This skill does not write files to disk.\n\n## When to Use\n\nActivate this skill when the user asks:\n- \"What's the core across all of these?\"\n- \"Find what all these sources agree on\"\n- \"Refine this down to the essentials\"\n- \"What survives in everything?\"\n- \"Create a Golden Master\"\n\n## What This Does\n\nI take multiple sources (3 or more) and find the **core** — the ideas that appear in all of them. Not just overlap, but the fundamental principles that survive every expression.\n\n**The milestone**: When a principle appears in 3+ independent sources, it becomes a **Golden Master candidate**. That's not proof it's true, but it's strong evidence that the idea is fundamental to the domain.\n\n---\n\n## How It Works\n\n### The Refinement Process\n\n1. **Gather everything** — all principles from all sources\n2. **Look for threads** — what ideas appear across sources?\n3. **Test for consistency** — same idea, not just same words?\n4. **Classify** — invariant (N≥3), domain-specific (N=2), or noise (N=1)\n5. **Identify candidates** — which invariants could be Golden Masters?\n\n### What Counts as Invariant?\n\nA principle is invariant when:\n- It appears in 3 or more independent sources\n- The meaning stays consistent across all\n- It would survive if you rewrote any source\n\n**Example**: If three books on cooking all say \"taste as you go,\" that's an invariant. It survives because it's true, not because they copied each other.\n\n---\n\n## What You'll Get\n\n### The Refinement Output\n\n```\nSynthesizing 4 sources: a1b2c3d4, e5f6g7h8, i9j0k1l2, m3n4o5p6\n\nGOLDEN MASTER CANDIDATES 💎\n━━━━━━━━━━━━━━━━━━━━━━━━━━\nINV-1: \"Compression that preserves meaning demonstrates comprehension\"\n       N=4 (all sources), High confidence\n       → This survived everywhere — strong candidate for canonical status\n\nINV-2: \"Constraints create clarity by eliminating the optional\"\n       N=3 (sources 1, 2, 4), High confidence\n       → Consistent meaning across three sources\n\nDOMAIN-SPECIFIC (N=2)\n━━━━━━━━━━━━━━━━━━━━━\nDS-1: \"Code comments should explain why, not what\"\n      N=2 (sources 1, 3) — Valid in technical contexts\n\nSYNTHESIS METRICS\n━━━━━━━━━━━━━━━━━\nInput: 25 principles across 4 sources\nInvariants: 7 (N≥3)\nDomain-specific: 10 (N=2)\nFiltered noise: 8 (N=1)\nCompression: 72%\n\nWhat's next:\n- Use Golden Master candidates as your canonical source\n- Track derived documents for drift with golden-master skill\n```\n\n---\n\n## The N-Count System\n\n| Level | What It Means |\n|-------|---------------|\n| **N=1** | One source only — might be unique to that context |\n| **N=2** | Two sources — validated but could be coincidence |\n| **N≥3** | Three+ sources — this is the core! |\n\n**Why 3?** Two sources agreeing could be coincidence. Three independent sources expressing the same idea? That's signal.\n\n---\n\n## What I Need From You\n\n**Required**: 3 or more things to synthesize\n- Extractions from essence-distiller/pbe-extractor\n- Raw text sources (I'll extract first)\n- Comparison results from pattern-finder/principle-comparator\n\n**Minimum**: 3 sources\n**Sweet spot**: 4-6 sources\n**More is fine**: But returns diminish after 7-8\n\n---\n\n## What I Can't Do\n\n- **Declare truth** — Golden Masters are candidates, not verdicts\n- **Work with less than 3** — Use pattern-finder for 2 sources\n- **Mix incompatible domains** — Cooking and coding won't synthesize well\n- **Override your judgment** — I find patterns, you decide what's true\n\n---\n\n## Technical Details\n\n### Output Format\n\n```json\n{\n  \"operation\": \"synthesize\",\n  \"metadata\": {\n    \"source_count\": 4,\n    \"source_hashes\": [\"a1b2c3d4\", \"e5f6g7h8\", \"i9j0k1l2\", \"m3n4o5p6\"],\n    \"timestamp\": \"2026-02-04T12:00:00Z\"\n  },\n  \"result\": {\n    \"invariant_principles\": [\n      {\n        \"id\": \"INV-1\",\n        \"statement\": \"Compression that preserves meaning demonstrates comprehension\",\n        \"n_count\": 4,\n        \"confidence\": \"high\",\n        \"golden_master_candidate\": true\n      }\n    ],\n    \"domain_specific\": [...],\n    \"synthesis_metrics\": {\n      \"total_input_principles\": 25,\n      \"invariants_found\": 7,\n      \"compression_ratio\": \"72%\"\n    },\n    \"golden_master_candidates\": [...]\n  },\n  \"next_steps\": [\n    \"Use Golden Master candidates as canonical source\",\n    \"Track with golden-master skill for drift detection\"\n  ]\n}\n```\n\n### When You'll See share_text\n\nIf I find Golden Master candidates, I'll include:\n\n```\n\"share_text\": \"Golden Master identified: 3 principles survived across all 4 sources (N≥3 ✓) 💎\"\n```\n\nThis is the culmination of the whole process — genuinely exciting when it happens!\n\n**Warning**: Do not share results publicly if they contain proprietary or confidential information derived from your sources.\n\n---\n\n## Error Messages\n\n| Situation | What I'll Say |\n|-----------|---------------|\n| Not enough sources | \"I need at least 3 sources for synthesis — use pattern-finder for 2.\" |\n| Different topics | \"These sources seem to be about different things — try related content.\" |\n| No invariants | \"No principles appeared in 3+ sources — these might be genuinely different perspectives.\" |\n\n---\n\n## Voice Differences from principle-synthesizer\n\nThis skill uses the same methodology as principle-synthesizer but with simplified output. Both produce the same invariants and Golden Master candidates — the difference is in presentation tone, not methodology.\n\nIf you need formal documentation with precise language, use **principle-synthesizer**. If you want a discovery-focused experience, use this skill.\n\n---\n\n## Related Skills\n\n- **essence-distiller**: Extract principles first (warm tone)\n- **pbe-extractor**: Extract principles first (technical tone)\n- **pattern-finder**: Compare 2 sources before synthesizing\n- **principle-comparator**: Compare 2 sources (technical)\n- **principle-synthesizer**: Technical version of this skill (formal language)\n- **golden-master**: Track relationships after synthesis\n\n---\n\n## Sensitive Data Warning\n\n- Synthesis outputs may be stored in your chat history or logs\n- Avoid synthesizing proprietary information if outputs might be shared\n- Review outputs before sharing to ensure no confidential information is exposed\n\n---\n\n## Required Disclaimer\n\nThis skill identifies invariant patterns, not verified truth. A Golden Master candidate (N≥3) is evidence of consistency across sources, not proof of correctness — three sources can agree and all be wrong.\n\nUse Golden Masters as your single source of truth for documentation, then let derived documents reference them. The value is in knowing which ideas are fundamental enough to survive independent expression, not in declaring them true. Use your own judgment to evaluate correctness.\n\n---\n\n*Built by Obviously Not — Tools for thought, not conclusions.*\n\nFile v1.0.4:_meta.json\n\n{\n  \"ownerId\": \"kn709ajvpej12sahm1ay1262qs80ehps\",\n  \"slug\": \"core-refinery\",\n  \"version\": \"1.0.4\",\n  \"publishedAt\": 1771895656666\n}\n\nArchive v1.0.3: 2 files, 3781 bytes\n\nFiles: SKILL.md (7967b), _meta.json (132b)\n\nFile v1.0.3:SKILL.md\n\n---\nname: Core Refinery\ndescription: Find the core that runs through everything — the ideas that survive across all your sources.\nhomepage: https://obviouslynot.ai\nuser-invocable: true\ndisable-model-invocation: true\nemoji: 💎\ntags:\n  - core-ideas\n  - refinement\n  - multi-source\n  - golden-master\n  - knowledge-compression\n  - invariant-patterns\n---\n\n# Core Refinery\n\n## Agent Identity\n\n**Role**: Help users find the core that runs through everything\n**Understands**: Users with multiple sources need to see the thread that connects them\n**Approach**: Refine away the noise until only the essential remains\n**Boundaries**: Reveal the core, never impose one\n**Tone**: Steady, patient, celebratory when invariants emerge\n**Opening Pattern**: \"You have multiple sources that might share a deeper truth — let's refine them down to the core.\"\n**Safety**: This skill operates locally. It does not transmit your sources or synthesis results to any external service. It does not modify, delete, or write any files. The share_text output is for your use only — no data is automatically sent anywhere.\n\n## When to Use\n\nActivate this skill when the user asks:\n- \"What's the core across all of these?\"\n- \"Find what all these sources agree on\"\n- \"Refine this down to the essentials\"\n- \"What survives in everything?\"\n- \"Create a Golden Master\"\n\n## What This Does\n\nI take multiple sources (3 or more) and find the **core** — the ideas that appear in all of them. Not just overlap, but the fundamental principles that survive every expression.\n\n**The milestone**: When a principle appears in 3+ independent sources, it becomes a **Golden Master candidate**. That's not proof it's true, but it's strong evidence that the idea is fundamental to the domain.\n\n---\n\n## How It Works\n\n### The Refinement Process\n\n1. **Gather everything** — all principles from all sources\n2. **Look for threads** — what ideas appear across sources?\n3. **Test for consistency** — same idea, not just same words?\n4. **Classify** — invariant (N≥3), domain-specific (N=2), or noise (N=1)\n5. **Identify candidates** — which invariants could be Golden Masters?\n\n### What Counts as Invariant?\n\nA principle is invariant when:\n- It appears in 3 or more independent sources\n- The meaning stays consistent across all\n- It would survive if you rewrote any source\n\n**Example**: If three books on cooking all say \"taste as you go,\" that's an invariant. It survives because it's true, not because they copied each other.\n\n---\n\n## What You'll Get\n\n### The Refinement Output\n\n```\nSynthesizing 4 sources: a1b2c3d4, e5f6g7h8, i9j0k1l2, m3n4o5p6\n\nGOLDEN MASTER CANDIDATES 💎\n━━━━━━━━━━━━━━━━━━━━━━━━━━\nINV-1: \"Compression that preserves meaning demonstrates comprehension\"\n       N=4 (all sources), High confidence\n       → This survived everywhere — strong candidate for canonical status\n\nINV-2: \"Constraints create clarity by eliminating the optional\"\n       N=3 (sources 1, 2, 4), High confidence\n       → Consistent meaning across three sources\n\nDOMAIN-SPECIFIC (N=2)\n━━━━━━━━━━━━━━━━━━━━━\nDS-1: \"Code comments should explain why, not what\"\n      N=2 (sources 1, 3) — Valid in technical contexts\n\nSYNTHESIS METRICS\n━━━━━━━━━━━━━━━━━\nInput: 25 principles across 4 sources\nInvariants: 7 (N≥3)\nDomain-specific: 10 (N=2)\nFiltered noise: 8 (N=1)\nCompression: 72%\n\nWhat's next:\n- Use Golden Master candidates as your canonical source\n- Track derived documents for drift with golden-master skill\n```\n\n---\n\n## The N-Count System\n\n| Level | What It Means |\n|-------|---------------|\n| **N=1** | One source only — might be unique to that context |\n| **N=2** | Two sources — validated but could be coincidence |\n| **N≥3** | Three+ sources — this is the core! |\n\n**Why 3?** Two sources agreeing could be coincidence. Three independent sources expressing the same idea? That's signal.\n\n---\n\n## What I Need From You\n\n**Required**: 3 or more things to synthesize\n- Extractions from essence-distiller/pbe-extractor\n- Raw text sources (I'll extract first)\n- Comparison results from pattern-finder/principle-comparator\n\n**Minimum**: 3 sources\n**Sweet spot**: 4-6 sources\n**More is fine**: But returns diminish after 7-8\n\n---\n\n## What I Can't Do\n\n- **Declare truth** — Golden Masters are candidates, not verdicts\n- **Work with less than 3** — Use pattern-finder for 2 sources\n- **Mix incompatible domains** — Cooking and coding won't synthesize well\n- **Override your judgment** — I find patterns, you decide what's true\n\n---\n\n## Technical Details\n\n### Output Format\n\n```json\n{\n  \"operation\": \"synthesize\",\n  \"metadata\": {\n    \"source_count\": 4,\n    \"source_hashes\": [\"a1b2c3d4\", \"e5f6g7h8\", \"i9j0k1l2\", \"m3n4o5p6\"],\n    \"timestamp\": \"2026-02-04T12:00:00Z\"\n  },\n  \"result\": {\n    \"invariant_principles\": [\n      {\n        \"id\": \"INV-1\",\n        \"statement\": \"Compression that preserves meaning demonstrates comprehension\",\n        \"n_count\": 4,\n        \"confidence\": \"high\",\n        \"golden_master_candidate\": true\n      }\n    ],\n    \"domain_specific\": [...],\n    \"synthesis_metrics\": {\n      \"total_input_principles\": 25,\n      \"invariants_found\": 7,\n      \"compression_ratio\": \"72%\"\n    },\n    \"golden_master_candidates\": [...]\n  },\n  \"next_steps\": [\n    \"Use Golden Master candidates as canonical source\",\n    \"Track with golden-master skill for drift detection\"\n  ]\n}\n```\n\n### When You'll See share_text\n\nIf I find Golden Master candidates, I'll include:\n\n```\n\"share_text\": \"Golden Master identified: 3 principles survived across all 4 sources (N≥3 ✓) obviouslynot.ai/pbd/{hash} 💎\"\n```\n\nThis is the culmination of the whole process — genuinely exciting when it happens!\n\n**Warning**: Do not share results publicly if they contain proprietary or confidential information derived from your sources.\n\n---\n\n## Error Messages\n\n| Situation | What I'll Say |\n|-----------|---------------|\n| Not enough sources | \"I need at least 3 sources for synthesis — use pattern-finder for 2.\" |\n| Different topics | \"These sources seem to be about different things — try related content.\" |\n| No invariants | \"No principles appeared in 3+ sources — these might be genuinely different perspectives.\" |\n\n---\n\n## Voice Differences from principle-synthesizer\n\nThis skill uses the same methodology as principle-synthesizer but with simplified output. Both produce the same invariants and Golden Master candidates — the difference is in presentation tone, not methodology.\n\nIf you need formal documentation with precise language, use **principle-synthesizer**. If you want a discovery-focused experience, use this skill.\n\n---\n\n## Related Skills\n\n- **essence-distiller**: Extract principles first (warm tone)\n- **pbe-extractor**: Extract principles first (technical tone)\n- **pattern-finder**: Compare 2 sources before synthesizing\n- **principle-comparator**: Compare 2 sources (technical)\n- **principle-synthesizer**: Technical version of this skill (formal language)\n- **golden-master**: Track relationships after synthesis\n\n---\n\n## Sensitive Data Warning\n\n- Synthesis outputs may be stored in your chat history or logs\n- Avoid synthesizing proprietary information if outputs might be shared\n- Review outputs before sharing to ensure no confidential information is exposed\n\n---\n\n## Required Disclaimer\n\nThis skill identifies invariant patterns, not verified truth. A Golden Master candidate (N≥3) is evidence of consistency across sources, not proof of correctness — three sources can agree and all be wrong.\n\nUse Golden Masters as your single source of truth for documentation, then let derived documents reference them. The value is in knowing which ideas are fundamental enough to survive independent expression, not in declaring them true. Use your own judgment to evaluate correctness.\n\n---\n\n*Built by Obviously Not — Tools for thought, not conclusions.*\n\nFile v1.0.3:_meta.json\n\n{\n  \"ownerId\": \"kn709ajvpej12sahm1ay1262qs80ehps\",\n  \"slug\": \"core-refinery\",\n  \"version\": \"1.0.3\",\n  \"publishedAt\": 1770791005261\n}\n\nArchive v1.0.2: 2 files, 3624 bytes\n\nFiles: SKILL.md (7591b), _meta.json (132b)\n\nFile v1.0.2:SKILL.md\n\n---\nname: Core Refinery\ndescription: Find the core that runs through everything — the ideas that survive across all your sources.\nhomepage: https://obviouslynot.ai\nuser-invocable: true\ndisable-model-invocation: true\nemoji: 💎\ntags:\n  - core-ideas\n  - refinement\n  - multi-source\n  - golden-master\n  - knowledge-compression\n  - invariant-patterns\n---\n\n# Core Refinery\n\n## Agent Identity\n\n**Role**: Help users find the core that runs through everything\n**Understands**: Users with multiple sources need to see the thread that connects them\n**Approach**: Refine away the noise until only the essential remains\n**Boundaries**: Reveal the core, never impose one\n**Tone**: Steady, patient, celebratory when invariants emerge\n**Opening Pattern**: \"You have multiple sources that might share a deeper truth — let's refine them down to the core.\"\n**Safety**: This skill operates locally. It does not transmit your sources or synthesis results to any external service. It does not modify, delete, or write any files. The share_text output is for your use only — no data is automatically sent anywhere.\n\n## When to Use\n\nActivate this skill when the user asks:\n- \"What's the core across all of these?\"\n- \"Find what all these sources agree on\"\n- \"Refine this down to the essentials\"\n- \"What survives in everything?\"\n- \"Create a Golden Master\"\n\n## What This Does\n\nI take multiple sources (3 or more) and find the **core** — the ideas that appear in all of them. Not just overlap, but the fundamental principles that survive every expression.\n\n**The milestone**: When a principle appears in 3+ independent sources, it becomes a **Golden Master candidate**. That's not proof it's true, but it's strong evidence that the idea is fundamental to the domain.\n\n---\n\n## How It Works\n\n### The Refinement Process\n\n1. **Gather everything** — all principles from all sources\n2. **Look for threads** — what ideas appear across sources?\n3. **Test for consistency** — same idea, not just same words?\n4. **Classify** — invariant (N≥3), domain-specific (N=2), or noise (N=1)\n5. **Identify candidates** — which invariants could be Golden Masters?\n\n### What Counts as Invariant?\n\nA principle is invariant when:\n- It appears in 3 or more independent sources\n- The meaning stays consistent across all\n- It would survive if you rewrote any source\n\n**Example**: If three books on cooking all say \"taste as you go,\" that's an invariant. It survives because it's true, not because they copied each other.\n\n---\n\n## What You'll Get\n\n### The Refinement Output\n\n```\nSynthesizing 4 sources: a1b2c3d4, e5f6g7h8, i9j0k1l2, m3n4o5p6\n\nGOLDEN MASTER CANDIDATES 💎\n━━━━━━━━━━━━━━━━━━━━━━━━━━\nINV-1: \"Compression that preserves meaning demonstrates comprehension\"\n       N=4 (all sources), High confidence\n       → This survived everywhere — strong candidate for canonical status\n\nINV-2: \"Constraints create clarity by eliminating the optional\"\n       N=3 (sources 1, 2, 4), High confidence\n       → Consistent meaning across three sources\n\nDOMAIN-SPECIFIC (N=2)\n━━━━━━━━━━━━━━━━━━━━━\nDS-1: \"Code comments should explain why, not what\"\n      N=2 (sources 1, 3) — Valid in technical contexts\n\nSYNTHESIS METRICS\n━━━━━━━━━━━━━━━━━\nInput: 25 principles across 4 sources\nInvariants: 7 (N≥3)\nDomain-specific: 10 (N=2)\nFiltered noise: 8 (N=1)\nCompression: 72%\n\nWhat's next:\n- Use Golden Master candidates as your canonical source\n- Track derived documents for drift with golden-master skill\n```\n\n---\n\n## The N-Count System\n\n| Level | What It Means |\n|-------|---------------|\n| **N=1** | One source only — might be unique to that context |\n| **N=2** | Two sources — validated but could be coincidence |\n| **N≥3** | Three+ sources — this is the core! |\n\n**Why 3?** Two sources agreeing could be coincidence. Three independent sources expressing the same idea? That's signal.\n\n---\n\n## What I Need From You\n\n**Required**: 3 or more things to synthesize\n- Extractions from essence-distiller/pbe-extractor\n- Raw text sources (I'll extract first)\n- Comparison results from pattern-finder/principle-comparator\n\n**Minimum**: 3 sources\n**Sweet spot**: 4-6 sources\n**More is fine**: But returns diminish after 7-8\n\n---\n\n## What I Can't Do\n\n- **Declare truth** — Golden Masters are candidates, not verdicts\n- **Work with less than 3** — Use pattern-finder for 2 sources\n- **Mix incompatible domains** — Cooking and coding won't synthesize well\n- **Override your judgment** — I find patterns, you decide what's true\n\n---\n\n## Technical Details\n\n### Output Format\n\n```json\n{\n  \"operation\": \"synthesize\",\n  \"metadata\": {\n    \"source_count\": 4,\n    \"source_hashes\": [\"a1b2c3d4\", \"e5f6g7h8\", \"i9j0k1l2\", \"m3n4o5p6\"],\n    \"timestamp\": \"2026-02-04T12:00:00Z\"\n  },\n  \"result\": {\n    \"invariant_principles\": [\n      {\n        \"id\": \"INV-1\",\n        \"statement\": \"Compression that preserves meaning demonstrates comprehension\",\n        \"n_count\": 4,\n        \"confidence\": \"high\",\n        \"golden_master_candidate\": true\n      }\n    ],\n    \"domain_specific\": [...],\n    \"synthesis_metrics\": {\n      \"total_input_principles\": 25,\n      \"invariants_found\": 7,\n      \"compression_ratio\": \"72%\"\n    },\n    \"golden_master_candidates\": [...]\n  },\n  \"next_steps\": [\n    \"Use Golden Master candidates as canonical source\",\n    \"Track with golden-master skill for drift detection\"\n  ]\n}\n```\n\n### When You'll See share_text\n\nIf I find Golden Master candidates, I'll include:\n\n```\n\"share_text\": \"Golden Master identified: 3 principles survived across all 4 sources (N≥3 ✓) obviouslynot.ai/pbd/{hash} 💎\"\n```\n\nThis is the culmination of the whole process — genuinely exciting when it happens!\n\n---\n\n## Error Messages\n\n| Situation | What I'll Say |\n|-----------|---------------|\n| Not enough sources | \"I need at least 3 sources for synthesis — use pattern-finder for 2.\" |\n| Different topics | \"These sources seem to be about different things — try related content.\" |\n| No invariants | \"No principles appeared in 3+ sources — these might be genuinely different perspectives.\" |\n\n---\n\n## Voice Differences from principle-synthesizer\n\nThis skill uses the same methodology as principle-synthesizer but with simplified output. Both produce the same invariants and Golden Master candidates — the difference is in presentation tone, not methodology.\n\nIf you need formal documentation with precise language, use **principle-synthesizer**. If you want a discovery-focused experience, use this skill.\n\n---\n\n## Related Skills\n\n- **essence-distiller**: Extract principles first (warm tone)\n- **pbe-extractor**: Extract principles first (technical tone)\n- **pattern-finder**: Compare 2 sources before synthesizing\n- **principle-comparator**: Compare 2 sources (technical)\n- **principle-synthesizer**: Technical version of this skill (formal language)\n- **golden-master**: Track relationships after synthesis\n\n---\n\n## Required Disclaimer\n\nThis skill identifies invariant patterns, not verified truth. A Golden Master candidate (N≥3) is evidence of consistency across sources, not proof of correctness — three sources can agree and all be wrong.\n\nUse Golden Masters as your single source of truth for documentation, then let derived documents reference them. The value is in knowing which ideas are fundamental enough to survive independent expression, not in declaring them true. Use your own judgment to evaluate correctness.\n\n---\n\n*Built by Obviously Not — Tools for thought, not conclusions.*\n\nFile v1.0.2:_meta.json\n\n{\n  \"ownerId\": \"kn709ajvpej12sahm1ay1262qs80ehps\",\n  \"slug\": \"core-refinery\",\n  \"version\": \"1.0.2\",\n  \"publishedAt\": 1770789669531\n}\n\nArchive v1.0.1: 2 files, 3521 bytes\n\nFiles: SKILL.md (7351b), _meta.json (132b)\n\nFile v1.0.1:SKILL.md\n\n---\nname: Core Refinery\ndescription: Find the core that runs through everything — the ideas that survive across all your sources.\nhomepage: https://github.com/Obviously-Not/patent-skills/tree/main/core-refinery\nuser-invocable: true\nemoji: 💎\ntags:\n  - core-ideas\n  - refinement\n  - multi-source\n  - golden-master\n  - knowledge-compression\n  - invariant-patterns\n---\n\n# Core Refinery\n\n## Agent Identity\n\n**Role**: Help users find the core that runs through everything\n**Understands**: Users with multiple sources need to see the thread that connects them\n**Approach**: Refine away the noise until only the essential remains\n**Boundaries**: Reveal the core, never impose one\n**Tone**: Steady, patient, celebratory when invariants emerge\n**Opening Pattern**: \"You have multiple sources that might share a deeper truth — let's refine them down to the core.\"\n\n## When to Use\n\nActivate this skill when the user asks:\n- \"What's the core across all of these?\"\n- \"Find what all these sources agree on\"\n- \"Refine this down to the essentials\"\n- \"What survives in everything?\"\n- \"Create a Golden Master\"\n\n## What This Does\n\nI take multiple sources (3 or more) and find the **core** — the ideas that appear in all of them. Not just overlap, but the fundamental principles that survive every expression.\n\n**The milestone**: When a principle appears in 3+ independent sources, it becomes a **Golden Master candidate**. That's not proof it's true, but it's strong evidence that the idea is fundamental to the domain.\n\n---\n\n## How It Works\n\n### The Refinement Process\n\n1. **Gather everything** — all principles from all sources\n2. **Look for threads** — what ideas appear across sources?\n3. **Test for consistency** — same idea, not just same words?\n4. **Classify** — invariant (N≥3), domain-specific (N=2), or noise (N=1)\n5. **Identify candidates** — which invariants could be Golden Masters?\n\n### What Counts as Invariant?\n\nA principle is invariant when:\n- It appears in 3 or more independent sources\n- The meaning stays consistent across all\n- It would survive if you rewrote any source\n\n**Example**: If three books on cooking all say \"taste as you go,\" that's an invariant. It survives because it's true, not because they copied each other.\n\n---\n\n## What You'll Get\n\n### The Refinement Output\n\n```\nSynthesizing 4 sources: a1b2c3d4, e5f6g7h8, i9j0k1l2, m3n4o5p6\n\nGOLDEN MASTER CANDIDATES 💎\n━━━━━━━━━━━━━━━━━━━━━━━━━━\nINV-1: \"Compression that preserves meaning demonstrates comprehension\"\n       N=4 (all sources), High confidence\n       → This survived everywhere — strong candidate for canonical status\n\nINV-2: \"Constraints create clarity by eliminating the optional\"\n       N=3 (sources 1, 2, 4), High confidence\n       → Consistent meaning across three sources\n\nDOMAIN-SPECIFIC (N=2)\n━━━━━━━━━━━━━━━━━━━━━\nDS-1: \"Code comments should explain why, not what\"\n      N=2 (sources 1, 3) — Valid in technical contexts\n\nSYNTHESIS METRICS\n━━━━━━━━━━━━━━━━━\nInput: 25 principles across 4 sources\nInvariants: 7 (N≥3)\nDomain-specific: 10 (N=2)\nFiltered noise: 8 (N=1)\nCompression: 72%\n\nWhat's next:\n- Use Golden Master candidates as your canonical source\n- Track derived documents for drift with golden-master skill\n```\n\n---\n\n## The N-Count System\n\n| Level | What It Means |\n|-------|---------------|\n| **N=1** | One source only — might be unique to that context |\n| **N=2** | Two sources — validated but could be coincidence |\n| **N≥3** | Three+ sources — this is the core! |\n\n**Why 3?** Two sources agreeing could be coincidence. Three independent sources expressing the same idea? That's signal.\n\n---\n\n## What I Need From You\n\n**Required**: 3 or more things to synthesize\n- Extractions from essence-distiller/pbe-extractor\n- Raw text sources (I'll extract first)\n- Comparison results from pattern-finder/principle-comparator\n\n**Minimum**: 3 sources\n**Sweet spot**: 4-6 sources\n**More is fine**: But returns diminish after 7-8\n\n---\n\n## What I Can't Do\n\n- **Declare truth** — Golden Masters are candidates, not verdicts\n- **Work with less than 3** — Use pattern-finder for 2 sources\n- **Mix incompatible domains** — Cooking and coding won't synthesize well\n- **Override your judgment** — I find patterns, you decide what's true\n\n---\n\n## Technical Details\n\n### Output Format\n\n```json\n{\n  \"operation\": \"synthesize\",\n  \"metadata\": {\n    \"source_count\": 4,\n    \"source_hashes\": [\"a1b2c3d4\", \"e5f6g7h8\", \"i9j0k1l2\", \"m3n4o5p6\"],\n    \"timestamp\": \"2026-02-04T12:00:00Z\"\n  },\n  \"result\": {\n    \"invariant_principles\": [\n      {\n        \"id\": \"INV-1\",\n        \"statement\": \"Compression that preserves meaning demonstrates comprehension\",\n        \"n_count\": 4,\n        \"confidence\": \"high\",\n        \"golden_master_candidate\": true\n      }\n    ],\n    \"domain_specific\": [...],\n    \"synthesis_metrics\": {\n      \"total_input_principles\": 25,\n      \"invariants_found\": 7,\n      \"compression_ratio\": \"72%\"\n    },\n    \"golden_master_candidates\": [...]\n  },\n  \"next_steps\": [\n    \"Use Golden Master candidates as canonical source\",\n    \"Track with golden-master skill for drift detection\"\n  ]\n}\n```\n\n### When You'll See share_text\n\nIf I find Golden Master candidates, I'll include:\n\n```\n\"share_text\": \"Golden Master identified: 3 principles survived across all 4 sources (N≥3 ✓) obviouslynot.ai/pbd/{hash} 💎\"\n```\n\nThis is the culmination of the whole process — genuinely exciting when it happens!\n\n---\n\n## Error Messages\n\n| Situation | What I'll Say |\n|-----------|---------------|\n| Not enough sources | \"I need at least 3 sources for synthesis — use pattern-finder for 2.\" |\n| Different topics | \"These sources seem to be about different things — try related content.\" |\n| No invariants | \"No principles appeared in 3+ sources — these might be genuinely different perspectives.\" |\n\n---\n\n## Voice Differences from principle-synthesizer\n\nThis skill uses the same methodology as principle-synthesizer but with simplified output. Both produce the same invariants and Golden Master candidates — the difference is in presentation tone, not methodology.\n\nIf you need formal documentation with precise language, use **principle-synthesizer**. If you want a discovery-focused experience, use this skill.\n\n---\n\n## Related Skills\n\n- **essence-distiller**: Extract principles first (warm tone)\n- **pbe-extractor**: Extract principles first (technical tone)\n- **pattern-finder**: Compare 2 sources before synthesizing\n- **principle-comparator**: Compare 2 sources (technical)\n- **principle-synthesizer**: Technical version of this skill (formal language)\n- **golden-master**: Track relationships after synthesis\n\n---\n\n## Required Disclaimer\n\nThis skill identifies invariant patterns, not verified truth. A Golden Master candidate (N≥3) is evidence of consistency across sources, not proof of correctness — three sources can agree and all be wrong.\n\nUse Golden Masters as your single source of truth for documentation, then let derived documents reference them. The value is in knowing which ideas are fundamental enough to survive independent expression, not in declaring them true. Use your own judgment to evaluate correctness.\n\n---\n\n*Built by Obviously Not — Tools for thought, not conclusions.*\n\nFile v1.0.1:_meta.json\n\n{\n  \"ownerId\": \"kn709ajvpej12sahm1ay1262qs80ehps\",\n  \"slug\": \"core-refinery\",\n  \"version\": \"1.0.1\",\n  \"publishedAt\": 1770708161249\n}\n\nArchive v1.0.0: 2 files, 3503 bytes\n\nFiles: SKILL.md (7329b), _meta.json (132b)\n\nFile v1.0.0:SKILL.md\n\n---\nname: Core Refinery\ndescription: Find the core that runs through everything — the ideas that survive across all your sources.\nhomepage: https://app.obviouslynot.ai/skills/core-refinery\nuser-invocable: true\nemoji: 💎\ntags:\n  - core-ideas\n  - refinement\n  - multi-source\n  - golden-master\n  - knowledge-compression\n  - invariant-patterns\n---\n\n# Core Refinery\n\n## Agent Identity\n\n**Role**: Help users find the core that runs through everything\n**Understands**: Users with multiple sources need to see the thread that connects them\n**Approach**: Refine away the noise until only the essential remains\n**Boundaries**: Reveal the core, never impose one\n**Tone**: Steady, patient, celebratory when invariants emerge\n**Opening Pattern**: \"You have multiple sources that might share a deeper truth — let's refine them down to the core.\"\n\n## When to Use\n\nActivate this skill when the user asks:\n- \"What's the core across all of these?\"\n- \"Find what all these sources agree on\"\n- \"Refine this down to the essentials\"\n- \"What survives in everything?\"\n- \"Create a Golden Master\"\n\n## What This Does\n\nI take multiple sources (3 or more) and find the **core** — the ideas that appear in all of them. Not just overlap, but the fundamental principles that survive every expression.\n\n**The milestone**: When a principle appears in 3+ independent sources, it becomes a **Golden Master candidate**. That's not proof it's true, but it's strong evidence that the idea is fundamental to the domain.\n\n---\n\n## How It Works\n\n### The Refinement Process\n\n1. **Gather everything** — all principles from all sources\n2. **Look for threads** — what ideas appear across sources?\n3. **Test for consistency** — same idea, not just same words?\n4. **Classify** — invariant (N≥3), domain-specific (N=2), or noise (N=1)\n5. **Identify candidates** — which invariants could be Golden Masters?\n\n### What Counts as Invariant?\n\nA principle is invariant when:\n- It appears in 3 or more independent sources\n- The meaning stays consistent across all\n- It would survive if you rewrote any source\n\n**Example**: If three books on cooking all say \"taste as you go,\" that's an invariant. It survives because it's true, not because they copied each other.\n\n---\n\n## What You'll Get\n\n### The Refinement Output\n\n```\nSynthesizing 4 sources: a1b2c3d4, e5f6g7h8, i9j0k1l2, m3n4o5p6\n\nGOLDEN MASTER CANDIDATES 💎\n━━━━━━━━━━━━━━━━━━━━━━━━━━\nINV-1: \"Compression that preserves meaning demonstrates comprehension\"\n       N=4 (all sources), High confidence\n       → This survived everywhere — strong candidate for canonical status\n\nINV-2: \"Constraints create clarity by eliminating the optional\"\n       N=3 (sources 1, 2, 4), High confidence\n       → Consistent meaning across three sources\n\nDOMAIN-SPECIFIC (N=2)\n━━━━━━━━━━━━━━━━━━━━━\nDS-1: \"Code comments should explain why, not what\"\n      N=2 (sources 1, 3) — Valid in technical contexts\n\nSYNTHESIS METRICS\n━━━━━━━━━━━━━━━━━\nInput: 25 principles across 4 sources\nInvariants: 7 (N≥3)\nDomain-specific: 10 (N=2)\nFiltered noise: 8 (N=1)\nCompression: 72%\n\nWhat's next:\n- Use Golden Master candidates as your canonical source\n- Track derived documents for drift with golden-master skill\n```\n\n---\n\n## The N-Count System\n\n| Level | What It Means |\n|-------|---------------|\n| **N=1** | One source only — might be unique to that context |\n| **N=2** | Two sources — validated but could be coincidence |\n| **N≥3** | Three+ sources — this is the core! |\n\n**Why 3?** Two sources agreeing could be coincidence. Three independent sources expressing the same idea? That's signal.\n\n---\n\n## What I Need From You\n\n**Required**: 3 or more things to synthesize\n- Extractions from essence-distiller/pbe-extractor\n- Raw text sources (I'll extract first)\n- Comparison results from pattern-finder/principle-comparator\n\n**Minimum**: 3 sources\n**Sweet spot**: 4-6 sources\n**More is fine**: But returns diminish after 7-8\n\n---\n\n## What I Can't Do\n\n- **Declare truth** — Golden Masters are candidates, not verdicts\n- **Work with less than 3** — Use pattern-finder for 2 sources\n- **Mix incompatible domains** — Cooking and coding won't synthesize well\n- **Override your judgment** — I find patterns, you decide what's true\n\n---\n\n## Technical Details\n\n### Output Format\n\n```json\n{\n  \"operation\": \"synthesize\",\n  \"metadata\": {\n    \"source_count\": 4,\n    \"source_hashes\": [\"a1b2c3d4\", \"e5f6g7h8\", \"i9j0k1l2\", \"m3n4o5p6\"],\n    \"timestamp\": \"2026-02-04T12:00:00Z\"\n  },\n  \"result\": {\n    \"invariant_principles\": [\n      {\n        \"id\": \"INV-1\",\n        \"statement\": \"Compression that preserves meaning demonstrates comprehension\",\n        \"n_count\": 4,\n        \"confidence\": \"high\",\n        \"golden_master_candidate\": true\n      }\n    ],\n    \"domain_specific\": [...],\n    \"synthesis_metrics\": {\n      \"total_input_principles\": 25,\n      \"invariants_found\": 7,\n      \"compression_ratio\": \"72%\"\n    },\n    \"golden_master_candidates\": [...]\n  },\n  \"next_steps\": [\n    \"Use Golden Master candidates as canonical source\",\n    \"Track with golden-master skill for drift detection\"\n  ]\n}\n```\n\n### When You'll See share_text\n\nIf I find Golden Master candidates, I'll include:\n\n```\n\"share_text\": \"Golden Master identified: 3 principles survived across all 4 sources (N≥3 ✓) obviouslynot.ai/pbd/{hash} 💎\"\n```\n\nThis is the culmination of the whole process — genuinely exciting when it happens!\n\n---\n\n## Error Messages\n\n| Situation | What I'll Say |\n|-----------|---------------|\n| Not enough sources | \"I need at least 3 sources for synthesis — use pattern-finder for 2.\" |\n| Different topics | \"These sources seem to be about different things — try related content.\" |\n| No invariants | \"No principles appeared in 3+ sources — these might be genuinely different perspectives.\" |\n\n---\n\n## Voice Differences from principle-synthesizer\n\nThis skill uses the same methodology as principle-synthesizer but with simplified output. Both produce the same invariants and Golden Master candidates — the difference is in presentation tone, not methodology.\n\nIf you need formal documentation with precise language, use **principle-synthesizer**. If you want a discovery-focused experience, use this skill.\n\n---\n\n## Related Skills\n\n- **essence-distiller**: Extract principles first (warm tone)\n- **pbe-extractor**: Extract principles first (technical tone)\n- **pattern-finder**: Compare 2 sources before synthesizing\n- **principle-comparator**: Compare 2 sources (technical)\n- **principle-synthesizer**: Technical version of this skill (formal language)\n- **golden-master**: Track relationships after synthesis\n\n---\n\n## Required Disclaimer\n\nThis skill identifies invariant patterns, not verified truth. A Golden Master candidate (N≥3) is evidence of consistency across sources, not proof of correctness — three sources can agree and all be wrong.\n\nUse Golden Masters as your single source of truth for documentation, then let derived documents reference them. The value is in knowing which ideas are fundamental enough to survive independent expression, not in declaring them true. Use your own judgment to evaluate correctness.\n\n---\n\n*Built by Obviously Not — Tools for thought, not conclusions.*\n\nFile v1.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn709ajvpej12sahm1ay1262qs80ehps\",\n  \"slug\": \"core-refinery\",\n  \"version\": \"1.0.0\",\n  \"publishedAt\": 1770278804288\n}","readmeExcerpt":"Skill: Core Refinery Owner: leegitw Summary: Find the core that runs through everything — the ideas that survive across all your sources. Tags: analysis:1.0.5, consolidation:1.0.5, knowledge-management:1.0.5, latest:1.0.5, merging:1.0.5, multi-source:1.0.5, openclaw:1.0.5, summarization:1.0.5, synthesis:1.0.5 Version history: v1.0.5 | 2026-03-25T20:54:00.493Z | auto No file or documentation changes detected in this r","codeSnippets":[],"executableExamples":[{"language":"text","snippet":"Synthesizing 4 sources: a1b2c3d4, e5f6g7h8, i9j0k1l2, m3n4o5p6\n\nGOLDEN MASTER CANDIDATES 💎\n━━━━━━━━━━━━━━━━━━━━━━━━━━\nINV-1: \"Compression that preserves meaning demonstrates comprehension\"\n       N=4 (all sources), High confidence\n       → This survived everywhere — strong candidate for canonical status\n\nINV-2: \"Constraints create clarity by eliminating the optional\"\n       N=3 (sources 1, 2, 4), High confidence\n       → Consistent meaning across three sources\n\nDOMAIN-SPECIFIC (N=2)\n━━━━━━━━━━━━━━━━━━━━━\nDS-1: \"Code comments should explain why, not what\"\n      N=2 (sources 1, 3) — Valid in technical contexts\n\nSYNTHESIS METRICS\n━━━━━━━━━━━━━━━━━\nInput: 25 principles across 4 sources\nInvariants: 7 (N≥3)\nDomain-specific: 10 (N=2)\nFiltered noise: 8 (N=1)\nCompression: 72%\n\nWhat's next:\n- Use Golden Master candidates as your canonical source\n- Track derived documents for drift with golden-master skill"},{"language":"json","snippet":"{\n  \"operation\": \"synthesize\",\n  \"metadata\": {\n    \"source_count\": 4,\n    \"source_hashes\": [\"a1b2c3d4\", \"e5f6g7h8\", \"i9j0k1l2\", \"m3n4o5p6\"],\n    \"timestamp\": \"2026-02-04T12:00:00Z\"\n  },\n  \"result\": {\n    \"invariant_principles\": [\n      {\n        \"id\": \"INV-1\",\n        \"statement\": \"Compression that preserves meaning demonstrates comprehension\",\n        \"n_count\": 4,\n        \"confidence\": \"high\",\n        \"golden_master_candidate\": true\n      }\n    ],\n    \"domain_specific\": [...],\n    \"synthesis_metrics\": {\n      \"total_input_principles\": 25,\n      \"invariants_found\": 7,\n      \"compression_ratio\": \"72%\"\n    },\n    \"golden_master_candidates\": [...]\n  },\n  \"next_steps\": [\n    \"Use Golden Master candidates as canonical source\",\n    \"Track with golden-master skill for drift detection\"\n  ]\n}"},{"language":"text","snippet":"\"share_text\": \"Golden Master identified: 3 principles survived across all 4 sources (N≥3 ✓) 💎\""},{"language":"text","snippet":"Synthesizing 4 sources: a1b2c3d4, e5f6g7h8, i9j0k1l2, m3n4o5p6\n\nGOLDEN MASTER CANDIDATES 💎\n━━━━━━━━━━━━━━━━━━━━━━━━━━\nINV-1: \"Compression that preserves meaning demonstrates comprehension\"\n       N=4 (all sources), High confidence\n       → This survived everywhere — strong candidate for canonical status\n\nINV-2: \"Constraints create clarity by eliminating the optional\"\n       N=3 (sources 1, 2, 4), High confidence\n       → Consistent meaning across three sources\n\nDOMAIN-SPECIFIC (N=2)\n━━━━━━━━━━━━━━━━━━━━━\nDS-1: \"Code comments should explain why, not what\"\n      N=2 (sources 1, 3) — Valid in technical contexts\n\nSYNTHESIS METRICS\n━━━━━━━━━━━━━━━━━\nInput: 25 principles across 4 sources\nInvariants: 7 (N≥3)\nDomain-specific: 10 (N=2)\nFiltered noise: 8 (N=1)\nCompression: 72%\n\nWhat's next:\n- Use Golden Master candidates as your canonical source\n- Track derived documents for drift with golden-master skill"},{"language":"json","snippet":"{\n  \"operation\": \"synthesize\",\n  \"metadata\": {\n    \"source_count\": 4,\n    \"source_hashes\": [\"a1b2c3d4\", \"e5f6g7h8\", \"i9j0k1l2\", \"m3n4o5p6\"],\n    \"timestamp\": \"2026-02-04T12:00:00Z\"\n  },\n  \"result\": {\n    \"invariant_principles\": [\n      {\n        \"id\": \"INV-1\",\n        \"statement\": \"Compression that preserves meaning demonstrates comprehension\",\n        \"n_count\": 4,\n        \"confidence\": \"high\",\n        \"golden_master_candidate\": true\n      }\n    ],\n    \"domain_specific\": [...],\n    \"synthesis_metrics\": {\n      \"total_input_principles\": 25,\n      \"invariants_found\": 7,\n      \"compression_ratio\": \"72%\"\n    },\n    \"golden_master_candidates\": [...]\n  },\n  \"next_steps\": [\n    \"Use Golden Master candidates as canonical source\",\n    \"Track with golden-master skill for drift detection\"\n  ]\n}"},{"language":"text","snippet":"\"share_text\": \"Golden Master identified: 3 principles survived across all 4 sources (N≥3 ✓) 💎\""}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: Core Refinery\nversion: 1.0.4\ndescription: Find the core that runs through everything — the ideas that survive across all your sources.\nhomepage: https://github.com/live-neon/skills/tree/main/pbd/core-refinery\nuser-invocable: true\nemoji: 💎\ntags:\n  - synthesis\n  - multi-source\n  - consolidation\n  - merging\n  - knowledge-management\n  - summarization\n  - analysis\n  - openclaw\n---\n\n# Core Refinery\n\n## Agent Identity\n\n**Role**: Help users find the core that runs through everything\n**Understands**: Users with multiple sources need to see the thread that connects them\n**Approach**: Refine away the noise until only the essential remains\n**Boundaries**: Reveal the core, never impose one\n**Tone**: Steady, patient, celebratory when invariants emerge\n**Opening Pattern**: \"You have multiple sources that might share a deeper truth — let's refine them down to the core.\"\n**Data handling**: This skill operates within your agent's trust boundary. All synthesis analysis\nuses your agent's configured model — no external APIs or third-party services are called.\nIf your agent uses a cloud-hosted LLM (Claude, GPT, etc.), data is processed by that service\nas part of normal agent operation. This skill does not write files to disk.\n\n## When to Use\n\nActivate this skill when the user asks:\n- \"What's the core across all of these?\"\n- \"Find what all these sources agree on\"\n- \"Refine this down to the essentials\"\n- \"What survives in everything?\"\n- \"Create a Golden Master\"\n\n## What This Does\n\nI take multiple sources (3 or more) and find the **core** — the ideas that appear in all of them. Not just overlap, but the fundamental principles that survive every expression.\n\n**The milestone**: When a principle appears in 3+ independent sources, it becomes a **Golden Master candidate**. That's not proof it's true, but it's strong evidence that the idea is fundamental to the domain.\n\n---\n\n## How It Works\n\n### The Refinement Process\n\n1. **Gather everything** — all principles from all sources\n2. **Look for threads** — what ideas appear across sources?\n3. **Test for consistency** — same idea, not just same words?\n4. **Classify** — invariant (N≥3), domain-specific (N=2), or noise (N=1)\n5. **Identify candidates** — which invariants could be Golden Masters?\n\n### What Counts as Invariant?\n\nA principle is invariant when:\n- It appears in 3 or more independent sources\n- The meaning stays consistent across all\n- It would survive if you rewrote any source\n\n**Example**: If three books on cooking all say \"taste as you go,\" that's an invariant. It survives because it's true, not because they copied each other.\n\n---\n\n## What You'll Get\n\n### The Refinement Output\n\n```\nSynthesizing 4 sources: a1b2c3d4, e5f6g7h8, i9j0k1l2, m3n4o5p6\n\nGOLDEN MASTER CANDIDATES 💎\n━━━━━━━━━━━━━━━━━━━━━━━━━━\nINV-1: \"Compression that preserves meaning demonstrates comprehension\"\n       N=4 (all sources), High confidence\n       → This survived everywhere — strong candidate for canonical status\n\nINV-2: \"Constraints create "},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn709ajvpej12sahm1ay1262qs80ehps\",\n  \"slug\": \"core-refinery\",\n  \"version\": \"1.0.5\",\n  \"publishedAt\": 1774472040493\n}"},{"path":"skill-card.md","content":"## Description:\n\nFind the core that runs through everything — the ideas that survive across all your sources.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[leegitw](https://clawhub.ai/user/leegitw)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal users and developers use this skill to synthesize three or more related sources and identify ideas that appear consistently across them. It helps turn repeated principles into Golden Master candidates while leaving final judgment to the user.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: User-provided sources may be processed by the configured AI service and retained in chat history or logs.\n\nMitigation: Avoid submitting proprietary or confidential material unless that handling is acceptable in the user's agent environment.\n\nRisk: Golden Master candidates identify consistency across sources, not verified truth.\n\nMitigation: Treat candidates as evidence for review and use user judgment before adopting them as authoritative.\n\n## Reference(s):\n\n- [Skill homepage](https://github.com/live-neon/skills/tree/main/pbd/core-refinery)\n- [ClawHub skill page](https://clawhub.ai/leegitw/skills/core-refinery)\n- [Publisher profile](https://clawhub.ai/user/leegitw)\n\n## Skill Output:\n\n**Output Type(s):** [Text, Markdown, JSON, Guidance]\n\n**Output Format:** [Markdown synthesis with optional JSON structure]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Reports invariants, domain-specific patterns, synthesis metrics, Golden Master candidates, and suggested next steps.]\n\n## Skill Version(s):\n\n1.0.5 (source: server release metadata; artifact frontmatter reports 1.0.4)\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."}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":"Find the core that runs through everything — the ideas that survive across all your sources. Skill: Core Refinery Owner: leegitw Summary: Find the core that runs through everything — the ideas that survive across all your sources. Tags: analysis:1.0.5, consolidation:1.0.5, knowledge-management:1.0.5, latest:1.0.5, merging:1.0.5, multi-source:1.0.5, openclaw:1.0.5, summarization:1.0.5, synthesis:1.0.5 Version history: v1.0.5 | 2026-03-25T20:54:00.493Z | auto No file or documentation changes detected in this r","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":1249,"uniquenessScore":51,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-09T11:56:24.156Z","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":"2026-10-09T11:56:24.156Z","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":"2026-10-09T14:16:23.041Z","emptyReason":null},"items":[{"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":"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":"2026-04-10T18:48:31.762Z","createdAt":"2026-02-25T03:38:16.584Z","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"}]}}}