{"id":"f472a8f6-6e65-41c0-be35-47b4ac70dba4","entityType":"agent","slug":"clawhub-clarezoe-dev-inbox","name":"dev-inbox","canonicalUrl":"https://www.xpersona.co/agent/clawhub-clarezoe-dev-inbox","canonicalPath":"/agent/clawhub-clarezoe-dev-inbox","generatedAt":"2026-10-12T01:57:34.191Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-11T20:51:21.880Z","emptyReason":null},"description":"Triage and persist work outside the explicit active objective. Use at message intake when a message contains multiple requests, an item is unrelated or deferred, the user asks to remember, track, log, or handle something later, or invokes dev-inbox. After resume or context compaction, audit requested-but-unrecorded items before continuing. Skill: dev-inbox Owner: clarezoe Summary: Triage and persist work outside the explicit active objective. Use at message intake when a message contains multiple requests, an item is unrelated or deferred, the user asks to remember, track, log, or handle something later, or invokes dev-inbox. After resume or context compaction, audit requested-but-unrecorded items before continuing. Tags: issue-management:1.0.0, issue-","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 994 downloads reported by the source. Last updated 10/11/2026.","installCommand":"clawhub skill install s178sk484g47vwcbvvfcjpftfx8461fz:dev-inbox","sourceUrl":"https://clawhub.ai/clarezoe/dev-inbox","homepage":"https://clawhub.ai/clarezoe/skills/dev-inbox","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/clarezoe/dev-inbox","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/clarezoe/skills/dev-inbox","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":60,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"Triage and persist work outside the explicit active objective. Use at message intake when a message contains multiple requests, an item is unrelated or deferred"},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-11T20:51:21.880Z","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":"2026-10-11T20:51:21.880Z","emptyReason":null},"stars":null,"forks":null,"downloads":994,"likes":null,"task":null,"library":null,"packageName":null,"latestVersion":"1.0.3","tractionLabel":"994 downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T20:51:21.867Z","emptyReason":null},"lastUpdatedAt":"2026-10-11T20:51:21.880Z","lastCrawledAt":"2026-10-11T20:51:21.867Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-12T20:51:21.867Z","lastVerifiedAt":null,"highlights":[{"version":"1.0.3","createdAt":"2026-08-03T22:44:00.798Z","changelog":"Harden multi-topic intake routing, treat explicit record requests as consent, and recover requested-but-unrecorded items after resume or compaction.","fileCount":4,"zipByteSize":6713},{"version":"1.0.2","createdAt":"2026-06-09T13:21:00.854Z","changelog":"Add OpenCode, Kiro, Devin, Codex, Cursor, and Windsurf skill compatibility","fileCount":4,"zipByteSize":5642},{"version":"1.0.1","createdAt":"2026-06-09T13:08:21.927Z","changelog":"Normalize SKILL.md frontmatter for OpenClaw, Hermes, and Claude compatibility","fileCount":4,"zipByteSize":5587},{"version":"1.0.0","createdAt":"2026-05-26T22:47:15.552Z","changelog":"Initial publish: universal task triage and routing skill for bugs, features, improvements, and ideas","fileCount":3,"zipByteSize":4625}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s178sk484g47vwcbvvfcjpftfx8461fz:dev-inbox","setupComplexity":"low","setupSteps":["Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.","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-clarezoe-dev-inbox/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-dev-inbox/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-dev-inbox/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-dev-inbox/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-dev-inbox/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-dev-inbox/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-12T01:57:34.190Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-dev-inbox/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-dev-inbox/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-dev-inbox/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-clarezoe-dev-inbox/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":"2026-10-11T20:51:21.880Z","emptyReason":null},"readme":"Skill: dev-inbox\n\nOwner: clarezoe\n\nSummary: Triage and persist work outside the explicit active objective. Use at message intake when a message contains multiple requests, an item is unrelated or deferred, the user asks to remember, track, log, or handle something later, or invokes dev-inbox. After resume or context compaction, audit requested-but-unrecorded items before continuing.\n\nTags: issue-management:1.0.0, issue-tracking:1.0.3, latest:1.0.3, productivity:1.0.0, session-management:1.0.3, task-routing:1.0.0, task-triage:1.0.3, triage:1.0.0, workflow:1.0.3\n\nVersion history:\n\nv1.0.3 | 2026-08-03T22:44:00.798Z | user\n\nHarden multi-topic intake routing, treat explicit record requests as consent, and recover requested-but-unrecorded items after resume or compaction.\n\nv1.0.2 | 2026-06-09T13:21:00.854Z | user\n\nAdd OpenCode, Kiro, Devin, Codex, Cursor, and Windsurf skill compatibility\n\nv1.0.1 | 2026-06-09T13:08:21.927Z | user\n\nNormalize SKILL.md frontmatter for OpenClaw, Hermes, and Claude compatibility\n\nv1.0.0 | 2026-05-26T22:47:15.552Z | user\n\nInitial publish: universal task triage and routing skill for bugs, features, improvements, and ideas\n\nArchive index:\n\nArchive v1.0.3: 4 files, 6713 bytes\n\nFiles: README.md (1794b), skill-card.md (2057b), SKILL.md (9232b), _meta.json (128b)\n\nFile v1.0.3:SKILL.md\n\n---\nname: \"dev-inbox\"\ndescription: \"Triage and persist work outside the explicit active objective. Use at message intake when a message contains multiple requests, an item is unrelated or deferred, the user asks to remember, track, log, or handle something later, or invokes dev-inbox. After resume or context compaction, audit requested-but-unrecorded items before continuing.\"\nlicense: \"MIT\"\nargument-hint: \"Describe the problem, idea, or improvement\"\nallowed-tools: \"Bash, Read, Write, Grep, Glob\"\nmetadata: {\"version\":\"1.0.3\",\"category\":\"workflow\",\"triggers\":[\"记一下\",\"记下来\",\"以后再说\",\"这个先不管\",\"open an issue\",\"log this\",\"track this\",\"dev inbox\",\"inbox\",\"/dev-inbox\"],\"license\":\"MIT\",\"tags\":[\"workflow\",\"task-triage\",\"issue-tracking\",\"session-management\"],\"hermes\":{\"tags\":[\"workflow\",\"task-triage\",\"issue-tracking\",\"session-management\"]}}\n---\n\n# Dev Inbox\n\nTriage anything that comes up during a session — bugs, features, improvements, fleeting ideas — and route it to the right place so it is never lost and always discoverable by future sessions.\n\nThis skill works in any context: software development, writing, design, or any task.\n\n---\n\n## Intake Gate\n\nRun this gate before acting on every multi-request message:\n\n1. Identify the explicit active objective from the active task artifact or the user's latest scoped instruction.\n2. Split every bullet, sentence-level request, and aside into atomic items.\n3. Classify each item as active-objective work, a blocking prerequisite, or unrelated/deferred inbox work.\n4. Route explicit inbox items before returning to active-objective work.\n\nConversation history, open files, and `git diff` are context, not proof that an item belongs to the active objective. Mere presence in the current conversation does not make a topic current-task work.\n\nAfter a resume, handoff, or context compaction, audit the recovered context for any explicit recording request that lacks a confirmed destination. Persist it before continuing the active objective.\n\n---\n\n## Decision Tree\n\n```\nUser says something\n│\n├─ Is it part of the current task?\n│   (Current task = active task artifact or latest explicitly scoped objective)\n│   ├─ YES → Do it directly. Stop here.\n│   └─ NO ↓\n│\n├─ Does it block the current task?\n│   ├─ YES → Handle it first, then return to current task.\n│   └─ NO ↓\n│\n├─ Priority?\n│   ├─ HIGH — would cause data loss, money loss, security issue, or blocks others\n│   ├─ NORMAL — clearly needs doing, but not urgent\n│   └─ LOW — nice-to-have, fleeting thought, cosmetic\n│\n└─ Record it + ensure future discoverability\n```\n\n---\n\n## Classification\n\nAssign one type and one priority:\n\n| Type | Meaning | Example |\n|------|---------|---------|\n| `fix` | Something existing is broken or wrong | Bug, incorrect content, wrong behavior |\n| `add` | Something new is needed | Feature, new section, new capability |\n| `improve` | Works but could be better | Better wording, cleaner UI, performance |\n| `idea` | Fleeting thought, maybe later | \"What if we also...\" |\n\nPriority: `high` / `normal` / `low`\n\n---\n\n## Interaction Pattern\n\n### Explicit request\n\nTreat an explicit request to record or defer an item as consent. This includes phrases such as \"记一下\", \"以后再说\", \"log this\", \"track this\", and direct `dev-inbox` invocation.\n\n1. Classify the item.\n2. Check for an existing related record.\n3. Persist or merge it without asking for confirmation again.\n4. Report the record location, then return to the active objective.\n\nAsk one focused question only when deduplication is genuinely ambiguous.\n\n### Proactive inference\n\nWhen you identify something that doesn't belong to the current task, respond with a **concrete suggestion** — not an open question:\n\n> This doesn't seem related to the current task. I suggest recording it as:\n>\n> **Type**: fix | **Priority**: normal\n> **Title**: Receipt is not generated after order submission\n>\n> Confirm?\n\nThe user responds with one word (yes/no/adjust). Then execute.\n\n---\n\n## Regression Examples\n\n### Multi-topic explicit request\n\nUser:\n\n> 继续修 Browser UI。另外打开会话特别慢，用 Dev-inbox 记一下，后面再说。\n\nExpected order:\n\n1. Split the Browser UI work and session-opening performance problem into separate items.\n2. Deduplicate and persist the performance item first because the user explicitly requested recording.\n3. Report where it was recorded.\n4. Return to the Browser UI objective.\n\n### Resume recovery\n\nIf a resume summary says the user requested an item be recorded but no destination or confirmation exists, treat it as unresolved. Deduplicate and persist it before resuming the active objective.\n\n---\n\n## Where to Record\n\nDetect the environment and pick the best destination. The goal is **future discoverability** — the record must surface in a future session without the user remembering it exists.\n\n### Priority order:\n\n1. **GitHub remote + `gh` CLI available**\n   - Check: `git remote get-url origin 2>/dev/null && command -v gh`\n   - Action: `gh issue create` with title, label, and body\n   - Why discoverable: Agent can `gh issue list --state open --label <type>`\n   - Merge logic: Search open issues with same label + similar title keywords. If found → append checklist item. If unsure → ask user.\n\n2. **Agent memory system available** (Devin memories, Claude memory, etc.)\n   - Action: Write to memory with `[TODO]` prefix and type/priority metadata\n   - Why discoverable: Automatically loaded on next session start\n\n3. **Project directory exists**\n   - Action: Append to `TODO.md` in project root (create if absent)\n   - Format: Grouped by type, each item has priority tag\n   - Why discoverable: Agent should read `TODO.md` at session start\n\n4. **None of the above**\n   - Action: Output formatted content for the user to place manually\n   - Format: Ready to paste anywhere\n\n### After recording, confirm:\n\n```\nRecorded: [type] [title]\nLocation: [where it was saved]\nDiscovery: [how a future session will find it]\n```\n\n---\n\n## Record Templates\n\nAdapt detail level to priority — lower priority = lighter format.\n\n### fix (high/normal)\n\n```markdown\n## Problem\n[What is broken / wrong]\n\n## Expected\n[What should happen]\n\n## Context\n[Where/when discovered, related task if any]\n```\n\n### fix (low)\n\n```markdown\n- [One-line description of what's wrong]\n```\n\n### add\n\n```markdown\n## What\n[What to add]\n\n## Why\n[Why it matters]\n```\n\n### improve\n\n```markdown\n- [What it is now] → [What it should be]\n```\n\n### idea\n\n```markdown\n- [One sentence]\n```\n\n---\n\n## Merge Logic\n\nBefore creating a new record, check if a related one already exists:\n\n- **GitHub Issues**: `gh issue list --state open --label <type>` — scan titles for keyword overlap\n- **TODO.md**: Scan same type section for similar items\n- **Memory**: Check for `[TODO]` entries with similar content\n\nIf a match is found → append as a sub-item or checklist entry.\nIf uncertain → ask: \"This looks related to [existing item]. Add to it, or create separate?\"\n\n---\n\n## GitHub Issue Labels\n\nWhen using GitHub Issues, apply these labels (create if they don't exist):\n\n```bash\ngh label create \"fix\" --color \"d73a4a\" --description \"Something is broken\" 2>/dev/null\ngh label create \"add\" --color \"0075ca\" --description \"New feature or content\" 2>/dev/null\ngh label create \"improve\" --color \"a2eeef\" --description \"Enhancement to existing\" 2>/dev/null\ngh label create \"idea\" --color \"e4e669\" --description \"Exploration, maybe later\" 2>/dev/null\ngh label create \"high\" --color \"b60205\" --description \"High priority\" 2>/dev/null\ngh label create \"low\" --color \"c5def5\" --description \"Low priority\" 2>/dev/null\n```\n\n---\n\n## No-`gh` Fallback\n\nIf `gh` is not installed but a GitHub remote exists, output:\n\n```\nI can't create the issue automatically (gh CLI not found).\nHere's the issue ready to paste:\n\nTitle: [title]\nLabels: [type], [priority]\nBody:\n---\n[formatted body]\n---\n\nCreate it at: [repo URL]/issues/new\n```\n\n---\n\n## Anti-patterns\n\n- Do not interrupt the user's flow with long explanations. Keep triage to 2–3 lines.\n- Do not ask open-ended questions like \"What do you want to do with this?\" — propose a concrete action.\n- Do not ask for confirmation after an explicit recording or deferral request; the request is consent.\n- Do not create records that can't be found later. Always state the discovery path.\n- Do not duplicate: always check for existing related records first.\n- Do not use this skill for things that ARE part of the current task — just do them.\n- Do not treat every topic in the conversation, open files, or `git diff` as part of the active objective.\n- Do not resume active work while a recovered explicit recording request remains unresolved.\n- Do not generate a file and leave it disconnected. The record must be in a system the agent will check.\n\n---\n\n## Relationship to Other Skills\n\n- **`session-handoff`** / **`close-loop`**: At session end, mention count of items recorded this session (e.g., \"Recorded 3 items to dev-inbox this session\"). No need to re-save — items are already persisted.\n- **`handoff-receiver`**: When resuming, check for pending inbox items as part of context loading.\n\nFile v1.0.3:README.md\n\n# dev-inbox\n\nUniversal task triage and routing for AI agent sessions.\n\n## What it does\n\n`dev-inbox` triages anything that comes up during a session — bugs, features, improvements, fleeting ideas — and routes it to the right place so it is never lost and always discoverable by future sessions.\n\nWorks in any context: software development, writing, design, or any task.\n\n## When to use\n\n- A message contains multiple requests that need to be separated before work starts\n- You notice something unrelated to the current task (a bug, idea, or improvement)\n- User says \"记一下\", \"log this\", \"open an issue\", \"以后再说\", \"track this\"\n- A resumed or compacted session mentions a requested item that was never recorded\n\n## How it works\n\n1. **Atomize** — split multi-topic messages into active, blocking, and unrelated/deferred items\n2. **Classify** — assign type (`fix` / `add` / `improve` / `idea`) and priority (`high` / `normal` / `low`)\n3. **Route** — automatically pick the best destination (GitHub Issue / agent memory / TODO.md)\n4. **Recover** — after resume or compaction, persist requested-but-unrecorded items before continuing\n5. **Ensure discoverability** — every record includes how a future session will find it\n\n## Key features\n\n- Proactive: agent intervenes when it detects off-task items\n- Explicit requests: records after deduplication without asking for confirmation again\n- Proactive inference: proposes title, type, and priority for one-word confirmation\n- Merge logic: checks for existing related records before creating duplicates\n- Environment-adaptive: works with or without GitHub, with or without agent memory\n\n---\n\n**Source**: [github.com/Fei2-Labs/skill-genie](https://github.com/Fei2-Labs/skill-genie)\n**Author**: [@clarezoe](https://x.com/clarezoe)\n\nFile v1.0.3:_meta.json\n\n{\n  \"ownerId\": \"kn71ntcwxg33gv53ryabqt0vw981g9my\",\n  \"slug\": \"dev-inbox\",\n  \"version\": \"1.0.3\",\n  \"publishedAt\": 1785797040798\n}\n\nFile v1.0.3:skill-card.md\n\n## Description:\n\nTriage and persist work outside the explicit active objective when a message contains multiple requests, an item is unrelated or deferred, the user asks to remember, track, log, or handle something later, or invokes dev-inbox.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[clarezoe](https://clawhub.ai/user/clarezoe)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and agents use this skill to separate active work from deferred or unrelated items, classify those items, and persist them to a discoverable follow-up destination.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Session content may be saved or published to GitHub issues, agent memory, or TODO.md without enough destination-specific confirmation.\n\nMitigation: Review the workflow before installing and add explicit confirmation for durable or remote writes where required.\n\nRisk: Secrets, customer data, private vulnerability details, or sensitive internal plans could be persisted if they appear in deferred work items.\n\nMitigation: Avoid using the skill with sensitive content unless the workflow is modified to require redaction before records are written.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/clarezoe/skills/dev-inbox)\n\n## Skill Output:\n\n**Output Type(s):** [Text, Markdown, Shell commands, Configuration, Guidance]\n\n**Output Format:** [Markdown guidance with record templates, optional shell commands, and concise confirmation text]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May create or update GitHub issues, agent memory entries, or TODO.md records depending on the available environment.]\n\n## Skill Version(s):\n\n1.0.3 (source: server release evidence and skill metadata)\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 v1.0.2: 4 files, 5642 bytes\n\nFiles: README.md (1341b), skill-card.md (2143b), SKILL.md (7087b), _meta.json (128b)\n\nFile v1.0.2:SKILL.md\n\n---\nname: \"dev-inbox\"\ndescription: \"Triage and route anything that comes up during a session — bugs, ideas, improvements, small fixes — to the right place. Use when the user or you notice something that may not belong to the current task.\"\nlicense: \"MIT\"\nargument-hint: \"Describe the problem, idea, or improvement\"\nallowed-tools: \"Bash, Read, Write, Grep, Glob\"\nmetadata: {\"version\":\"1.0.2\",\"category\":\"workflow\",\"triggers\":[\"记一下\",\"记下来\",\"以后再说\",\"这个先不管\",\"open an issue\",\"log this\",\"track this\",\"dev inbox\",\"inbox\",\"/dev-inbox\"],\"license\":\"MIT\",\"tags\":[\"workflow\",\"task-triage\",\"issue-tracking\",\"session-management\"],\"hermes\":{\"tags\":[\"workflow\",\"task-triage\",\"issue-tracking\",\"session-management\"]}}\n---\n\n# Dev Inbox\n\nTriage anything that comes up during a session — bugs, features, improvements, fleeting ideas — and route it to the right place so it is never lost and always discoverable by future sessions.\n\nThis skill works in any context: software development, writing, design, or any task.\n\n---\n\n## When to Activate\n\n- **Proactive**: You notice the user said something unrelated to the current task (a bug, idea, improvement, or tangent). Intervene immediately with a suggestion.\n- **Reactive**: The user explicitly says \"记一下\", \"log this\", \"open an issue\", \"以后再说\", \"track this\", or invokes `dev-inbox`.\n\n---\n\n## Decision Tree\n\n```\nUser says something\n│\n├─ Is it part of the current task?\n│   (Current task = current conversation + open files + git diff)\n│   ├─ YES → Do it directly. Stop here.\n│   └─ NO ↓\n│\n├─ Does it block the current task?\n│   ├─ YES → Handle it first, then return to current task.\n│   └─ NO ↓\n│\n├─ Priority?\n│   ├─ HIGH — would cause data loss, money loss, security issue, or blocks others\n│   ├─ NORMAL — clearly needs doing, but not urgent\n│   └─ LOW — nice-to-have, fleeting thought, cosmetic\n│\n└─ Record it + ensure future discoverability\n```\n\n---\n\n## Classification\n\nAssign one type and one priority:\n\n| Type | Meaning | Example |\n|------|---------|---------|\n| `fix` | Something existing is broken or wrong | Bug, incorrect content, wrong behavior |\n| `add` | Something new is needed | Feature, new section, new capability |\n| `improve` | Works but could be better | Better wording, cleaner UI, performance |\n| `idea` | Fleeting thought, maybe later | \"What if we also...\" |\n\nPriority: `high` / `normal` / `low`\n\n---\n\n## Interaction Pattern\n\nWhen you identify something that doesn't belong to the current task, respond with a **concrete suggestion** — not an open question:\n\n> This doesn't seem related to the current task. I suggest recording it as:\n>\n> **Type**: fix | **Priority**: normal\n> **Title**: Receipt is not generated after order submission\n>\n> Confirm?\n\nThe user responds with one word (yes/no/adjust). Then execute.\n\n---\n\n## Where to Record\n\nDetect the environment and pick the best destination. The goal is **future discoverability** — the record must surface in a future session without the user remembering it exists.\n\n### Priority order:\n\n1. **GitHub remote + `gh` CLI available**\n   - Check: `git remote get-url origin 2>/dev/null && command -v gh`\n   - Action: `gh issue create` with title, label, and body\n   - Why discoverable: Agent can `gh issue list --state open --label <type>`\n   - Merge logic: Search open issues with same label + similar title keywords. If found → append checklist item. If unsure → ask user.\n\n2. **Agent memory system available** (Windsurf memories, Claude memory, etc.)\n   - Action: Write to memory with `[TODO]` prefix and type/priority metadata\n   - Why discoverable: Automatically loaded on next session start\n\n3. **Project directory exists**\n   - Action: Append to `TODO.md` in project root (create if absent)\n   - Format: Grouped by type, each item has priority tag\n   - Why discoverable: Agent should read `TODO.md` at session start\n\n4. **None of the above**\n   - Action: Output formatted content for the user to place manually\n   - Format: Ready to paste anywhere\n\n### After recording, confirm:\n\n```\nRecorded: [type] [title]\nLocation: [where it was saved]\nDiscovery: [how a future session will find it]\n```\n\n---\n\n## Record Templates\n\nAdapt detail level to priority — lower priority = lighter format.\n\n### fix (high/normal)\n\n```markdown\n## Problem\n[What is broken / wrong]\n\n## Expected\n[What should happen]\n\n## Context\n[Where/when discovered, related task if any]\n```\n\n### fix (low)\n\n```markdown\n- [One-line description of what's wrong]\n```\n\n### add\n\n```markdown\n## What\n[What to add]\n\n## Why\n[Why it matters]\n```\n\n### improve\n\n```markdown\n- [What it is now] → [What it should be]\n```\n\n### idea\n\n```markdown\n- [One sentence]\n```\n\n---\n\n## Merge Logic\n\nBefore creating a new record, check if a related one already exists:\n\n- **GitHub Issues**: `gh issue list --state open --label <type>` — scan titles for keyword overlap\n- **TODO.md**: Scan same type section for similar items\n- **Memory**: Check for `[TODO]` entries with similar content\n\nIf a match is found → append as a sub-item or checklist entry.\nIf uncertain → ask: \"This looks related to [existing item]. Add to it, or create separate?\"\n\n---\n\n## GitHub Issue Labels\n\nWhen using GitHub Issues, apply these labels (create if they don't exist):\n\n```bash\ngh label create \"fix\" --color \"d73a4a\" --description \"Something is broken\" 2>/dev/null\ngh label create \"add\" --color \"0075ca\" --description \"New feature or content\" 2>/dev/null\ngh label create \"improve\" --color \"a2eeef\" --description \"Enhancement to existing\" 2>/dev/null\ngh label create \"idea\" --color \"e4e669\" --description \"Exploration, maybe later\" 2>/dev/null\ngh label create \"high\" --color \"b60205\" --description \"High priority\" 2>/dev/null\ngh label create \"low\" --color \"c5def5\" --description \"Low priority\" 2>/dev/null\n```\n\n---\n\n## No-`gh` Fallback\n\nIf `gh` is not installed but a GitHub remote exists, output:\n\n```\nI can't create the issue automatically (gh CLI not found).\nHere's the issue ready to paste:\n\nTitle: [title]\nLabels: [type], [priority]\nBody:\n---\n[formatted body]\n---\n\nCreate it at: [repo URL]/issues/new\n```\n\n---\n\n## Anti-patterns\n\n- Do not interrupt the user's flow with long explanations. Keep triage to 2–3 lines.\n- Do not ask open-ended questions like \"What do you want to do with this?\" — propose a concrete action.\n- Do not create records that can't be found later. Always state the discovery path.\n- Do not duplicate: always check for existing related records first.\n- Do not use this skill for things that ARE part of the current task — just do them.\n- Do not generate a file and leave it disconnected. The record must be in a system the agent will check.\n\n---\n\n## Relationship to Other Skills\n\n- **`session-handoff`** / **`close-loop`**: At session end, mention count of items recorded this session (e.g., \"Recorded 3 items to dev-inbox this session\"). No need to re-save — items are already persisted.\n- **`handoff-receiver`**: When resuming, check for pending inbox items as part of context loading.\n\nFile v1.0.2:README.md\n\n# dev-inbox\n\nUniversal task triage and routing for AI agent sessions.\n\n## What it does\n\n`dev-inbox` triages anything that comes up during a session — bugs, features, improvements, fleeting ideas — and routes it to the right place so it is never lost and always discoverable by future sessions.\n\nWorks in any context: software development, writing, design, or any task.\n\n## When to use\n\n- You notice something unrelated to the current task (a bug, idea, or improvement)\n- User says \"记一下\", \"log this\", \"open an issue\", \"以后再说\", \"track this\"\n\n## How it works\n\n1. **Classify** — assign type (`fix` / `add` / `improve` / `idea`) and priority (`high` / `normal` / `low`)\n2. **Route** — automatically pick the best destination (GitHub Issue / agent memory / TODO.md)\n3. **Ensure discoverability** — every record includes how a future session will find it\n\n## Key features\n\n- Proactive: agent intervenes when it detects off-task items\n- Concrete suggestions: proposes title, type, priority — user confirms with one word\n- Merge logic: checks for existing related records before creating duplicates\n- Environment-adaptive: works with or without GitHub, with or without agent memory\n\n---\n\n**Source**: [github.com/Fei2-Labs/skill-genie](https://github.com/Fei2-Labs/skill-genie)\n**Author**: [@clarezoe](https://x.com/clarezoe)\n\nFile v1.0.2:_meta.json\n\n{\n  \"ownerId\": \"kn71ntcwxg33gv53ryabqt0vw981g9my\",\n  \"slug\": \"dev-inbox\",\n  \"version\": \"1.0.2\",\n  \"publishedAt\": 1781011260854\n}\n\nFile v1.0.2:skill-card.md\n\n## Description: <br>\nTriage and route anything that comes up during a session - bugs, ideas, improvements, small fixes - to the right place. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[clarezoe](https://clawhub.ai/user/clarezoe) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nDevelopers, writers, designers, and other agent users use this skill to capture off-task bugs, ideas, improvements, and follow-up work during a session, then route each item to GitHub Issues, agent memory, TODO.md, or a paste-ready fallback so it remains discoverable. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Broad automatic triggers may turn casual off-task phrases into durable notes, memory entries, or GitHub Issues. <br>\nMitigation: Configure the agent to ask for confirmation before writing records, especially before storing conversation content or creating GitHub Issues. <br>\nRisk: Persistent local or remote writes can preserve sensitive or irrelevant session content longer than intended. <br>\nMitigation: Review the proposed title, type, priority, destination, and body before approving any record. <br>\n\n\n## Reference(s): <br>\n- [ClawHub listing](https://clawhub.ai/clarezoe/dev-inbox) <br>\n- [README](artifact/README.md) <br>\n- [SKILL.md](artifact/SKILL.md) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [Guidance, Markdown, Shell commands, Configuration] <br>\n**Output Format:** [Markdown with inline shell commands and structured task records] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [May create or append persistent records in GitHub Issues, agent memory, TODO.md, or paste-ready text after user confirmation.] <br>\n\n## Skill Version(s): <br>\n1.0.2 (source: ClawHub release metadata and skill 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 v1.0.1: 4 files, 5587 bytes\n\nFiles: README.md (1341b), skill-card.md (2066b), SKILL.md (7087b), _meta.json (128b)\n\nFile v1.0.1:SKILL.md\n\n---\nname: \"dev-inbox\"\ndescription: \"Triage and route anything that comes up during a session — bugs, ideas, improvements, small fixes — to the right place. Use when the user or you notice something that may not belong to the current task.\"\nlicense: \"MIT\"\nargument-hint: \"Describe the problem, idea, or improvement\"\nallowed-tools: \"Bash, Read, Write, Grep, Glob\"\nmetadata: {\"version\":\"1.0.1\",\"category\":\"workflow\",\"triggers\":[\"记一下\",\"记下来\",\"以后再说\",\"这个先不管\",\"open an issue\",\"log this\",\"track this\",\"dev inbox\",\"inbox\",\"/dev-inbox\"],\"license\":\"MIT\",\"tags\":[\"workflow\",\"task-triage\",\"issue-tracking\",\"session-management\"],\"hermes\":{\"tags\":[\"workflow\",\"task-triage\",\"issue-tracking\",\"session-management\"]}}\n---\n\n# Dev Inbox\n\nTriage anything that comes up during a session — bugs, features, improvements, fleeting ideas — and route it to the right place so it is never lost and always discoverable by future sessions.\n\nThis skill works in any context: software development, writing, design, or any task.\n\n---\n\n## When to Activate\n\n- **Proactive**: You notice the user said something unrelated to the current task (a bug, idea, improvement, or tangent). Intervene immediately with a suggestion.\n- **Reactive**: The user explicitly says \"记一下\", \"log this\", \"open an issue\", \"以后再说\", \"track this\", or invokes `dev-inbox`.\n\n---\n\n## Decision Tree\n\n```\nUser says something\n│\n├─ Is it part of the current task?\n│   (Current task = current conversation + open files + git diff)\n│   ├─ YES → Do it directly. Stop here.\n│   └─ NO ↓\n│\n├─ Does it block the current task?\n│   ├─ YES → Handle it first, then return to current task.\n│   └─ NO ↓\n│\n├─ Priority?\n│   ├─ HIGH — would cause data loss, money loss, security issue, or blocks others\n│   ├─ NORMAL — clearly needs doing, but not urgent\n│   └─ LOW — nice-to-have, fleeting thought, cosmetic\n│\n└─ Record it + ensure future discoverability\n```\n\n---\n\n## Classification\n\nAssign one type and one priority:\n\n| Type | Meaning | Example |\n|------|---------|---------|\n| `fix` | Something existing is broken or wrong | Bug, incorrect content, wrong behavior |\n| `add` | Something new is needed | Feature, new section, new capability |\n| `improve` | Works but could be better | Better wording, cleaner UI, performance |\n| `idea` | Fleeting thought, maybe later | \"What if we also...\" |\n\nPriority: `high` / `normal` / `low`\n\n---\n\n## Interaction Pattern\n\nWhen you identify something that doesn't belong to the current task, respond with a **concrete suggestion** — not an open question:\n\n> This doesn't seem related to the current task. I suggest recording it as:\n>\n> **Type**: fix | **Priority**: normal\n> **Title**: Receipt is not generated after order submission\n>\n> Confirm?\n\nThe user responds with one word (yes/no/adjust). Then execute.\n\n---\n\n## Where to Record\n\nDetect the environment and pick the best destination. The goal is **future discoverability** — the record must surface in a future session without the user remembering it exists.\n\n### Priority order:\n\n1. **GitHub remote + `gh` CLI available**\n   - Check: `git remote get-url origin 2>/dev/null && command -v gh`\n   - Action: `gh issue create` with title, label, and body\n   - Why discoverable: Agent can `gh issue list --state open --label <type>`\n   - Merge logic: Search open issues with same label + similar title keywords. If found → append checklist item. If unsure → ask user.\n\n2. **Agent memory system available** (Windsurf memories, Claude memory, etc.)\n   - Action: Write to memory with `[TODO]` prefix and type/priority metadata\n   - Why discoverable: Automatically loaded on next session start\n\n3. **Project directory exists**\n   - Action: Append to `TODO.md` in project root (create if absent)\n   - Format: Grouped by type, each item has priority tag\n   - Why discoverable: Agent should read `TODO.md` at session start\n\n4. **None of the above**\n   - Action: Output formatted content for the user to place manually\n   - Format: Ready to paste anywhere\n\n### After recording, confirm:\n\n```\nRecorded: [type] [title]\nLocation: [where it was saved]\nDiscovery: [how a future session will find it]\n```\n\n---\n\n## Record Templates\n\nAdapt detail level to priority — lower priority = lighter format.\n\n### fix (high/normal)\n\n```markdown\n## Problem\n[What is broken / wrong]\n\n## Expected\n[What should happen]\n\n## Context\n[Where/when discovered, related task if any]\n```\n\n### fix (low)\n\n```markdown\n- [One-line description of what's wrong]\n```\n\n### add\n\n```markdown\n## What\n[What to add]\n\n## Why\n[Why it matters]\n```\n\n### improve\n\n```markdown\n- [What it is now] → [What it should be]\n```\n\n### idea\n\n```markdown\n- [One sentence]\n```\n\n---\n\n## Merge Logic\n\nBefore creating a new record, check if a related one already exists:\n\n- **GitHub Issues**: `gh issue list --state open --label <type>` — scan titles for keyword overlap\n- **TODO.md**: Scan same type section for similar items\n- **Memory**: Check for `[TODO]` entries with similar content\n\nIf a match is found → append as a sub-item or checklist entry.\nIf uncertain → ask: \"This looks related to [existing item]. Add to it, or create separate?\"\n\n---\n\n## GitHub Issue Labels\n\nWhen using GitHub Issues, apply these labels (create if they don't exist):\n\n```bash\ngh label create \"fix\" --color \"d73a4a\" --description \"Something is broken\" 2>/dev/null\ngh label create \"add\" --color \"0075ca\" --description \"New feature or content\" 2>/dev/null\ngh label create \"improve\" --color \"a2eeef\" --description \"Enhancement to existing\" 2>/dev/null\ngh label create \"idea\" --color \"e4e669\" --description \"Exploration, maybe later\" 2>/dev/null\ngh label create \"high\" --color \"b60205\" --description \"High priority\" 2>/dev/null\ngh label create \"low\" --color \"c5def5\" --description \"Low priority\" 2>/dev/null\n```\n\n---\n\n## No-`gh` Fallback\n\nIf `gh` is not installed but a GitHub remote exists, output:\n\n```\nI can't create the issue automatically (gh CLI not found).\nHere's the issue ready to paste:\n\nTitle: [title]\nLabels: [type], [priority]\nBody:\n---\n[formatted body]\n---\n\nCreate it at: [repo URL]/issues/new\n```\n\n---\n\n## Anti-patterns\n\n- Do not interrupt the user's flow with long explanations. Keep triage to 2–3 lines.\n- Do not ask open-ended questions like \"What do you want to do with this?\" — propose a concrete action.\n- Do not create records that can't be found later. Always state the discovery path.\n- Do not duplicate: always check for existing related records first.\n- Do not use this skill for things that ARE part of the current task — just do them.\n- Do not generate a file and leave it disconnected. The record must be in a system the agent will check.\n\n---\n\n## Relationship to Other Skills\n\n- **`session-handoff`** / **`close-loop`**: At session end, mention count of items recorded this session (e.g., \"Recorded 3 items to dev-inbox this session\"). No need to re-save — items are already persisted.\n- **`handoff-receiver`**: When resuming, check for pending inbox items as part of context loading.\n\nFile v1.0.1:README.md\n\n# dev-inbox\n\nUniversal task triage and routing for AI agent sessions.\n\n## What it does\n\n`dev-inbox` triages anything that comes up during a session — bugs, features, improvements, fleeting ideas — and routes it to the right place so it is never lost and always discoverable by future sessions.\n\nWorks in any context: software development, writing, design, or any task.\n\n## When to use\n\n- You notice something unrelated to the current task (a bug, idea, or improvement)\n- User says \"记一下\", \"log this\", \"open an issue\", \"以后再说\", \"track this\"\n\n## How it works\n\n1. **Classify** — assign type (`fix` / `add` / `improve` / `idea`) and priority (`high` / `normal` / `low`)\n2. **Route** — automatically pick the best destination (GitHub Issue / agent memory / TODO.md)\n3. **Ensure discoverability** — every record includes how a future session will find it\n\n## Key features\n\n- Proactive: agent intervenes when it detects off-task items\n- Concrete suggestions: proposes title, type, priority — user confirms with one word\n- Merge logic: checks for existing related records before creating duplicates\n- Environment-adaptive: works with or without GitHub, with or without agent memory\n\n---\n\n**Source**: [github.com/Fei2-Labs/skill-genie](https://github.com/Fei2-Labs/skill-genie)\n**Author**: [@clarezoe](https://x.com/clarezoe)\n\nFile v1.0.1:_meta.json\n\n{\n  \"ownerId\": \"kn71ntcwxg33gv53ryabqt0vw981g9my\",\n  \"slug\": \"dev-inbox\",\n  \"version\": \"1.0.1\",\n  \"publishedAt\": 1781010501927\n}\n\nFile v1.0.1:skill-card.md\n\n## Description: <br>\nTriage and route anything that comes up during a session, including bugs, ideas, improvements, and small fixes, to the right place. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[clarezoe](https://clawhub.ai/user/clarezoe) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nDevelopers and agent users use this skill to capture off-task bugs, features, improvements, and ideas during a session, then route them to GitHub issues, agent memory, TODO.md, or a paste-ready fallback so they remain discoverable. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Confirmed records may be written to public GitHub issues or labels when used in a public repository. <br>\nMitigation: Review the proposed title, body, labels, and destination before confirming, and avoid including confidential information. <br>\nRisk: The skill may persist follow-up work in GitHub issues, agent memory, or TODO.md. <br>\nMitigation: Install and activate it only when persistent task capture is desired, and verify the selected destination after recording. <br>\n\n\n## Reference(s): <br>\n- [Dev Inbox on ClawHub](https://clawhub.ai/clarezoe/dev-inbox) <br>\n- [Publisher profile](https://clawhub.ai/user/clarezoe) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [text, markdown, shell commands, guidance] <br>\n**Output Format:** [Markdown and plain text with optional shell commands] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [May create GitHub issues, memory entries, or TODO.md entries only after user confirmation and based on the detected environment.] <br>\n\n## Skill Version(s): <br>\n1.0.1 (source: server release metadata and SKILL.md 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 v1.0.0: 3 files, 4625 bytes\n\nFiles: skill-card.md (2010b), SKILL.md (6979b), _meta.json (128b)\n\nFile v1.0.0:SKILL.md\n\n---\nname: dev-inbox\ndescription: Triage and route anything that comes up during a session — bugs, ideas, improvements, small fixes — to the right place. Use when the user or you notice something that may not belong to the current task.\nargument-hint: \"Describe the problem, idea, or improvement\"\nlicense: MIT\nallowed-tools:\n  - Bash\n  - Read\n  - Write\n  - Grep\n  - Glob\nmetadata:\n  version: 1.0.0\n  category: workflow\n  triggers:\n    - \"记一下\"\n    - \"记下来\"\n    - \"以后再说\"\n    - \"这个先不管\"\n    - \"open an issue\"\n    - \"log this\"\n    - \"track this\"\n    - \"dev inbox\"\n    - \"inbox\"\n    - \"/dev-inbox\"\n---\n\n# Dev Inbox\n\nTriage anything that comes up during a session — bugs, features, improvements, fleeting ideas — and route it to the right place so it is never lost and always discoverable by future sessions.\n\nThis skill works in any context: software development, writing, design, or any task.\n\n---\n\n## When to Activate\n\n- **Proactive**: You notice the user said something unrelated to the current task (a bug, idea, improvement, or tangent). Intervene immediately with a suggestion.\n- **Reactive**: The user explicitly says \"记一下\", \"log this\", \"open an issue\", \"以后再说\", \"track this\", or invokes `dev-inbox`.\n\n---\n\n## Decision Tree\n\n```\nUser says something\n│\n├─ Is it part of the current task?\n│   (Current task = current conversation + open files + git diff)\n│   ├─ YES → Do it directly. Stop here.\n│   └─ NO ↓\n│\n├─ Does it block the current task?\n│   ├─ YES → Handle it first, then return to current task.\n│   └─ NO ↓\n│\n├─ Priority?\n│   ├─ HIGH — would cause data loss, money loss, security issue, or blocks others\n│   ├─ NORMAL — clearly needs doing, but not urgent\n│   └─ LOW — nice-to-have, fleeting thought, cosmetic\n│\n└─ Record it + ensure future discoverability\n```\n\n---\n\n## Classification\n\nAssign one type and one priority:\n\n| Type | Meaning | Example |\n|------|---------|---------|\n| `fix` | Something existing is broken or wrong | Bug, incorrect content, wrong behavior |\n| `add` | Something new is needed | Feature, new section, new capability |\n| `improve` | Works but could be better | Better wording, cleaner UI, performance |\n| `idea` | Fleeting thought, maybe later | \"What if we also...\" |\n\nPriority: `high` / `normal` / `low`\n\n---\n\n## Interaction Pattern\n\nWhen you identify something that doesn't belong to the current task, respond with a **concrete suggestion** — not an open question:\n\n> This doesn't seem related to the current task. I suggest recording it as:\n>\n> **Type**: fix | **Priority**: normal\n> **Title**: Receipt is not generated after order submission\n>\n> Confirm?\n\nThe user responds with one word (yes/no/adjust). Then execute.\n\n---\n\n## Where to Record\n\nDetect the environment and pick the best destination. The goal is **future discoverability** — the record must surface in a future session without the user remembering it exists.\n\n### Priority order:\n\n1. **GitHub remote + `gh` CLI available**\n   - Check: `git remote get-url origin 2>/dev/null && command -v gh`\n   - Action: `gh issue create` with title, label, and body\n   - Why discoverable: Agent can `gh issue list --state open --label <type>`\n   - Merge logic: Search open issues with same label + similar title keywords. If found → append checklist item. If unsure → ask user.\n\n2. **Agent memory system available** (Windsurf memories, Claude memory, etc.)\n   - Action: Write to memory with `[TODO]` prefix and type/priority metadata\n   - Why discoverable: Automatically loaded on next session start\n\n3. **Project directory exists**\n   - Action: Append to `TODO.md` in project root (create if absent)\n   - Format: Grouped by type, each item has priority tag\n   - Why discoverable: Agent should read `TODO.md` at session start\n\n4. **None of the above**\n   - Action: Output formatted content for the user to place manually\n   - Format: Ready to paste anywhere\n\n### After recording, confirm:\n\n```\nRecorded: [type] [title]\nLocation: [where it was saved]\nDiscovery: [how a future session will find it]\n```\n\n---\n\n## Record Templates\n\nAdapt detail level to priority — lower priority = lighter format.\n\n### fix (high/normal)\n\n```markdown\n## Problem\n[What is broken / wrong]\n\n## Expected\n[What should happen]\n\n## Context\n[Where/when discovered, related task if any]\n```\n\n### fix (low)\n\n```markdown\n- [One-line description of what's wrong]\n```\n\n### add\n\n```markdown\n## What\n[What to add]\n\n## Why\n[Why it matters]\n```\n\n### improve\n\n```markdown\n- [What it is now] → [What it should be]\n```\n\n### idea\n\n```markdown\n- [One sentence]\n```\n\n---\n\n## Merge Logic\n\nBefore creating a new record, check if a related one already exists:\n\n- **GitHub Issues**: `gh issue list --state open --label <type>` — scan titles for keyword overlap\n- **TODO.md**: Scan same type section for similar items\n- **Memory**: Check for `[TODO]` entries with similar content\n\nIf a match is found → append as a sub-item or checklist entry.\nIf uncertain → ask: \"This looks related to [existing item]. Add to it, or create separate?\"\n\n---\n\n## GitHub Issue Labels\n\nWhen using GitHub Issues, apply these labels (create if they don't exist):\n\n```bash\ngh label create \"fix\" --color \"d73a4a\" --description \"Something is broken\" 2>/dev/null\ngh label create \"add\" --color \"0075ca\" --description \"New feature or content\" 2>/dev/null\ngh label create \"improve\" --color \"a2eeef\" --description \"Enhancement to existing\" 2>/dev/null\ngh label create \"idea\" --color \"e4e669\" --description \"Exploration, maybe later\" 2>/dev/null\ngh label create \"high\" --color \"b60205\" --description \"High priority\" 2>/dev/null\ngh label create \"low\" --color \"c5def5\" --description \"Low priority\" 2>/dev/null\n```\n\n---\n\n## No-`gh` Fallback\n\nIf `gh` is not installed but a GitHub remote exists, output:\n\n```\nI can't create the issue automatically (gh CLI not found).\nHere's the issue ready to paste:\n\nTitle: [title]\nLabels: [type], [priority]\nBody:\n---\n[formatted body]\n---\n\nCreate it at: [repo URL]/issues/new\n```\n\n---\n\n## Anti-patterns\n\n- Do not interrupt the user's flow with long explanations. Keep triage to 2–3 lines.\n- Do not ask open-ended questions like \"What do you want to do with this?\" — propose a concrete action.\n- Do not create records that can't be found later. Always state the discovery path.\n- Do not duplicate: always check for existing related records first.\n- Do not use this skill for things that ARE part of the current task — just do them.\n- Do not generate a file and leave it disconnected. The record must be in a system the agent will check.\n\n---\n\n## Relationship to Other Skills\n\n- **`session-handoff`** / **`close-loop`**: At session end, mention count of items recorded this session (e.g., \"Recorded 3 items to dev-inbox this session\"). No need to re-save — items are already persisted.\n- **`handoff-receiver`**: When resuming, check for pending inbox items as part of context loading.\n\nFile v1.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn71ntcwxg33gv53ryabqt0vw981g9my\",\n  \"slug\": \"dev-inbox\",\n  \"version\": \"1.0.0\",\n  \"publishedAt\": 1779835635552\n}\n\nFile v1.0.0:skill-card.md\n\n## Description: <br>\nTriage and route anything that comes up during a session - bugs, ideas, improvements, small fixes - to the right place. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[clarezoe](https://clawhub.ai/user/clarezoe) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nDevelopers and agents use this skill to triage side tasks, bugs, improvements, and ideas that arise during a session and route them to a discoverable follow-up location. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: The skill can persist user-provided notes to GitHub issues, memory, or local TODO files. <br>\nMitigation: Confirm the destination and visibility before recording, especially for public GitHub repositories. <br>\nRisk: Broad activation phrases may cause the agent to suggest recording tangential or sensitive session content. <br>\nMitigation: Use the skill only for intentional follow-up tracking and avoid recording confidential or sensitive details unless persistence is acceptable. <br>\n\n\n## Reference(s): <br>\n- [Dev Inbox on ClawHub](https://clawhub.ai/clarezoe/dev-inbox) <br>\n- [Publisher profile: clarezoe](https://clawhub.ai/user/clarezoe) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [Text, Markdown, Shell commands, Files, Guidance] <br>\n**Output Format:** [Markdown and concise text with optional shell commands or file updates] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [May create GitHub issues, memory entries, or TODO.md items when the user confirms a record should be persisted.] <br>\n\n## Skill Version(s): <br>\n1.0.0 (source: server evidence and skill 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>","readmeExcerpt":"Skill: dev-inbox Owner: clarezoe Summary: Triage and persist work outside the explicit active objective. Use at message intake when a message contains multiple requests, an item is unrelated or deferred, the user asks to remember, track, log, or handle something later, or invokes dev-inbox. After resume or context compaction, audit requested-but-unrecorded items before continuing. Tags: issue-management:1.0.0, issue-","codeSnippets":[],"executableExamples":[{"language":"text","snippet":"User says something\n│\n├─ Is it part of the current task?\n│   (Current task = active task artifact or latest explicitly scoped objective)\n│   ├─ YES → Do it directly. Stop here.\n│   └─ NO ↓\n│\n├─ Does it block the current task?\n│   ├─ YES → Handle it first, then return to current task.\n│   └─ NO ↓\n│\n├─ Priority?\n│   ├─ HIGH — would cause data loss, money loss, security issue, or blocks others\n│   ├─ NORMAL — clearly needs doing, but not urgent\n│   └─ LOW — nice-to-have, fleeting thought, cosmetic\n│\n└─ Record it + ensure future discoverability"},{"language":"text","snippet":"Recorded: [type] [title]\nLocation: [where it was saved]\nDiscovery: [how a future session will find it]"},{"language":"markdown","snippet":"## Problem\n[What is broken / wrong]\n\n## Expected\n[What should happen]\n\n## Context\n[Where/when discovered, related task if any]"},{"language":"markdown","snippet":"- [One-line description of what's wrong]"},{"language":"markdown","snippet":"## What\n[What to add]\n\n## Why\n[Why it matters]"},{"language":"markdown","snippet":"- [What it is now] → [What it should be]"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: \"dev-inbox\"\ndescription: \"Triage and persist work outside the explicit active objective. Use at message intake when a message contains multiple requests, an item is unrelated or deferred, the user asks to remember, track, log, or handle something later, or invokes dev-inbox. After resume or context compaction, audit requested-but-unrecorded items before continuing.\"\nlicense: \"MIT\"\nargument-hint: \"Describe the problem, idea, or improvement\"\nallowed-tools: \"Bash, Read, Write, Grep, Glob\"\nmetadata: {\"version\":\"1.0.3\",\"category\":\"workflow\",\"triggers\":[\"记一下\",\"记下来\",\"以后再说\",\"这个先不管\",\"open an issue\",\"log this\",\"track this\",\"dev inbox\",\"inbox\",\"/dev-inbox\"],\"license\":\"MIT\",\"tags\":[\"workflow\",\"task-triage\",\"issue-tracking\",\"session-management\"],\"hermes\":{\"tags\":[\"workflow\",\"task-triage\",\"issue-tracking\",\"session-management\"]}}\n---\n\n# Dev Inbox\n\nTriage anything that comes up during a session — bugs, features, improvements, fleeting ideas — and route it to the right place so it is never lost and always discoverable by future sessions.\n\nThis skill works in any context: software development, writing, design, or any task.\n\n---\n\n## Intake Gate\n\nRun this gate before acting on every multi-request message:\n\n1. Identify the explicit active objective from the active task artifact or the user's latest scoped instruction.\n2. Split every bullet, sentence-level request, and aside into atomic items.\n3. Classify each item as active-objective work, a blocking prerequisite, or unrelated/deferred inbox work.\n4. Route explicit inbox items before returning to active-objective work.\n\nConversation history, open files, and `git diff` are context, not proof that an item belongs to the active objective. Mere presence in the current conversation does not make a topic current-task work.\n\nAfter a resume, handoff, or context compaction, audit the recovered context for any explicit recording request that lacks a confirmed destination. Persist it before continuing the active objective.\n\n---\n\n## Decision Tree\n\n```\nUser says something\n│\n├─ Is it part of the current task?\n│   (Current task = active task artifact or latest explicitly scoped objective)\n│   ├─ YES → Do it directly. Stop here.\n│   └─ NO ↓\n│\n├─ Does it block the current task?\n│   ├─ YES → Handle it first, then return to current task.\n│   └─ NO ↓\n│\n├─ Priority?\n│   ├─ HIGH — would cause data loss, money loss, security issue, or blocks others\n│   ├─ NORMAL — clearly needs doing, but not urgent\n│   └─ LOW — nice-to-have, fleeting thought, cosmetic\n│\n└─ Record it + ensure future discoverability\n```\n\n---\n\n## Classification\n\nAssign one type and one priority:\n\n| Type | Meaning | Example |\n|------|---------|---------|\n| `fix` | Something existing is broken or wrong | Bug, incorrect content, wrong behavior |\n| `add` | Something new is needed | Feature, new section, new capability |\n| `improve` | Works but could be better | Better wording, cleaner UI, performance |\n| `idea` | Fleeting thought, maybe later | \"What if we also...\" |\n\n"},{"path":"README.md","content":"# dev-inbox\n\nUniversal task triage and routing for AI agent sessions.\n\n## What it does\n\n`dev-inbox` triages anything that comes up during a session — bugs, features, improvements, fleeting ideas — and routes it to the right place so it is never lost and always discoverable by future sessions.\n\nWorks in any context: software development, writing, design, or any task.\n\n## When to use\n\n- A message contains multiple requests that need to be separated before work starts\n- You notice something unrelated to the current task (a bug, idea, or improvement)\n- User says \"记一下\", \"log this\", \"open an issue\", \"以后再说\", \"track this\"\n- A resumed or compacted session mentions a requested item that was never recorded\n\n## How it works\n\n1. **Atomize** — split multi-topic messages into active, blocking, and unrelated/deferred items\n2. **Classify** — assign type (`fix` / `add` / `improve` / `idea`) and priority (`high` / `normal` / `low`)\n3. **Route** — automatically pick the best destination (GitHub Issue / agent memory / TODO.md)\n4. **Recover** — after resume or compaction, persist requested-but-unrecorded items before continuing\n5. **Ensure discoverability** — every record includes how a future session will find it\n\n## Key features\n\n- Proactive: agent intervenes when it detects off-task items\n- Explicit requests: records after deduplication without asking for confirmation again\n- Proactive inference: proposes title, type, and priority for one-word confirmation\n- Merge logic: checks for existing related records before creating duplicates\n- Environment-adaptive: works with or without GitHub, with or without agent memory\n\n---\n\n**Source**: [github.com/Fei2-Labs/skill-genie](https://github.com/Fei2-Labs/skill-genie)\n**Author**: [@clarezoe](https://x.com/clarezoe)"},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn71ntcwxg33gv53ryabqt0vw981g9my\",\n  \"slug\": \"dev-inbox\",\n  \"version\": \"1.0.3\",\n  \"publishedAt\": 1785797040798\n}"},{"path":"skill-card.md","content":"## Description:\n\nTriage and persist work outside the explicit active objective when a message contains multiple requests, an item is unrelated or deferred, the user asks to remember, track, log, or handle something later, or invokes dev-inbox.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[clarezoe](https://clawhub.ai/user/clarezoe)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and agents use this skill to separate active work from deferred or unrelated items, classify those items, and persist them to a discoverable follow-up destination.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Session content may be saved or published to GitHub issues, agent memory, or TODO.md without enough destination-specific confirmation.\n\nMitigation: Review the workflow before installing and add explicit confirmation for durable or remote writes where required.\n\nRisk: Secrets, customer data, private vulnerability details, or sensitive internal plans could be persisted if they appear in deferred work items.\n\nMitigation: Avoid using the skill with sensitive content unless the workflow is modified to require redaction before records are written.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/clarezoe/skills/dev-inbox)\n\n## Skill Output:\n\n**Output Type(s):** [Text, Markdown, Shell commands, Configuration, Guidance]\n\n**Output Format:** [Markdown guidance with record templates, optional shell commands, and concise confirmation text]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May create or update GitHub issues, agent memory entries, or TODO.md records depending on the available environment.]\n\n## Skill Version(s):\n\n1.0.3 (source: server release evidence and skill metadata)\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":"Triage and persist work outside the explicit active objective. Use at message intake when a message contains multiple requests, an item is unrelated or deferred, the user asks to remember, track, log, or handle something later, or invokes dev-inbox. After resume or context compaction, audit requested-but-unrecorded items before continuing. Skill: dev-inbox Owner: clarezoe Summary: Triage and persist work outside the explicit active objective. Use at message intake when a message contains multiple requests, an item is unrelated or deferred, the user asks to remember, track, log, or handle something later, or invokes dev-inbox. After resume or context compaction, audit requested-but-unrecorded items before continuing. Tags: issue-management:1.0.0, issue-","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":1301,"uniquenessScore":46,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-11T20:51:21.880Z","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":"2026-10-11T20:51:21.880Z","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":"2026-10-12T01:57:34.191Z","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"}]}}}