{"id":"c5e87235-15ce-4278-b4af-46a97587079f","entityType":"agent","slug":"clawhub-englandtong-daily-workflow","name":"Daily Workflow","canonicalUrl":"https://www.xpersona.co/agent/clawhub-englandtong-daily-workflow","canonicalPath":"/agent/clawhub-englandtong-daily-workflow","generatedAt":"2026-10-10T23:47:03.515Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-10T21:03:35.764Z","emptyReason":null},"description":"Preserve evidence-backed project memory","descriptionLabel":"Source description","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 1.3K downloads reported by the source. Last updated 10/10/2026.","installCommand":"clawhub skill install s1712yyvmw7t8z7nqxpz1ydfad8540a4:daily-workflow","sourceUrl":"https://clawhub.ai/englandtong/daily-workflow","homepage":"https://clawhub.ai/englandtong/skills/daily-workflow","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/englandtong/daily-workflow","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/englandtong/skills/daily-workflow","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":62,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"Daily Workflow technical dossier on Xpersona with agent coverage, OPENCLEW support, and live trust metadata."},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-10T21:03:35.764Z","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":"2026-10-10T21:03:35.764Z","emptyReason":null},"stars":null,"forks":null,"downloads":1258,"packageName":null,"latestVersion":"4.1.0","tractionLabel":"1.3K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-10T21:03:35.740Z","emptyReason":null},"lastUpdatedAt":"2026-10-10T21:03:35.764Z","lastCrawledAt":"2026-10-10T21:03:35.740Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-11T21:03:35.740Z","lastVerifiedAt":null,"highlights":[{"version":"4.1.0","createdAt":"2026-09-07T17:06:14.694Z","changelog":"Rewrite the opening line to lead with the trigger phrase. Add QUICKSTART.md covering the four phrases and the two-file lightweight profile. Add memory-bloat control and backup-before-restructure rules.","fileCount":6,"zipByteSize":9620},{"version":"4.0.0","createdAt":"2026-08-14T15:02:56.272Z","changelog":"Add read-only orientation, one-writer state ownership, exact evidence outcomes, and safer resumable handoffs.","fileCount":5,"zipByteSize":7607},{"version":"3.0.0","createdAt":"2026-06-06T07:19:53.599Z","changelog":"A bilingual project memory workflow that records project metadata, targets, current status, completed work, pending work, next actions, archive history, and AI handoff notes at project start, checkpoint, wrap-up, or handoff. 一个中英双语项目记忆工作流 Skill，用于在项目启动、工作中段、收工或交接时保存项目元数据、目标、当前状态、已完成事项、待办事项、下一步计划、归档历史和 AI 交接信息。 Upgrade Introduction / 升级说明 This is the upgraded version of the previous Chinese and English Daily Workflow editions; going forward, both language editions will be maintained together in this single bilingual package. 这是此前中文与英文 Daily Workflow 两版的升级版；以后中英文版本会合并在这个双语包里统一维护。","fileCount":6,"zipByteSize":10004},{"version":"2.0.0","createdAt":"2026-05-09T15:19:53.510Z","changelog":"Version 2.0.0 introduces a project-local, context-compressing, bilingual workflow manager: - Adds context compression at checkpoint, wrap-up, and handoff, storing only essential info for user and next AI. - Unifies start/checkpoint/wrap-up/handoff flows with concise, actionable summaries instead of full transcripts. - Modernizes file structure: Docs/ directory with TARGET.md, STATUS.md, COMPLETED.md, PENDING.md, SCHEDULE.md, and optional HANDOFF.md. - Supports both Chinese and English trigger phrases and configurable workflow triggers in Docs/CONFIG.md. - Strongly focuses on clear AI-to-AI and user handoff, improved privacy, and reduced context bloat. - Includes initial project security policy (SECURITY.md) and new agent configuration file.","fileCount":5,"zipByteSize":6882},{"version":"1.1.0","createdAt":"2026-05-05T16:02:27.789Z","changelog":"### v1.1.0 (2026-05-05) ✨ 新功能• 新增「吃饭啦」触发点（Phase 1.5） — 午休前自动更新四个工作档案• 新增首次运行配置向导 — 支持自定义三个触发暗语 🔧 改进• 配置文件持久化到 .workbuddy/daily-workflow-config.json• 更新工作流图表，完整展示三个触发点流程• 优化文档模板，增加\"上午/下午\"时间段标识 🐛 修复• 修复收工时不更新 PROJECT_STATUS.md 的问题","fileCount":2,"zipByteSize":3688},{"version":"1.0.0","createdAt":"2026-05-04T04:23:17.746Z","changelog":"初始版本 1.0.0 功能介绍： - ✅ 支持「开工啦」触发词，自动读取项目状态 - ✅ 支持「收工啦」触发词，自动保存工作进度 - ✅ 自动管理 Docs/ 目录下的5个核心文档： - PROJECT_TARGET.md (项目目标) - PROJECT_STATUS.md (项目进度) - COMPLETED_JOBS.md (已完成工作) - PENDING_JOBS.md (未完成工作) - NEXT_STEPS.md (下一步计划) - ✅ 自动创建缺失的目录和文档（使用标准模板） - ✅ AI-to-AI 无缝交接，保留完整项目上下文 - ✅ 适用于多 AI 协作的项目管理场景 适用场景： - 每日工作开始/结束时的项目状态管理 - 多人/多 AI 协作时的上下文传递 - 需要详细工作日志的项目管理","fileCount":2,"zipByteSize":2830}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s1712yyvmw7t8z7nqxpz1ydfad8540a4:daily-workflow","setupComplexity":"low","setupSteps":["Install using `clawhub skill install s1712yyvmw7t8z7nqxpz1ydfad8540a4:daily-workflow` in an isolated environment before connecting it to live workloads.","No published capability contract is available yet, so validate auth and request/response behavior manually.","Review the upstream CLAWHUB listing at https://clawhub.ai/englandtong/daily-workflow before using production credentials."],"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-englandtong-daily-workflow/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-englandtong-daily-workflow/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-englandtong-daily-workflow/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-englandtong-daily-workflow/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-englandtong-daily-workflow/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-englandtong-daily-workflow/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-10T23:47:03.513Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-englandtong-daily-workflow/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-englandtong-daily-workflow/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-englandtong-daily-workflow/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-englandtong-daily-workflow/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":"medium","updatedAt":"2026-10-10T21:03:35.764Z","emptyReason":null},"readme":"Skill: Daily Workflow\n\nOwner: englandtong\n\nSummary: Preserve evidence-backed project memory\n\nTags: ai-handoff:4.1.0, ai-handover:4.1.0, ai-workflow:3.0.0, bilingual:4.1.0, checkpoint:4.1.0, chinese:4.1.0, context-compression:4.1.0, context-management:4.1.0, daily-routine:4.1.0, daily-standup:4.1.0, daily-workflow:4.1.0, docs:2.0.0, documentation:4.1.0, en:4.1.0, english:4.1.0, handoff:4.1.0, latest:4.1.0, productivity:4.1.0, project-management:4.1.0, project-memory:4.1.0, project-status:2.0.0, work-session:4.1.0, workflow:4.0.0, wrap-up:4.1.0, zh-cn:4.1.0\n\nVersion history:\n\nv4.1.0 | 2026-09-07T17:06:14.694Z | user\n\nRewrite the opening line to lead with the trigger phrase. Add QUICKSTART.md covering the four phrases and the two-file lightweight profile. Add memory-bloat control and backup-before-restructure rules.\n\nv4.0.0 | 2026-08-14T15:02:56.272Z | user\n\nAdd read-only orientation, one-writer state ownership, exact evidence outcomes, and safer resumable handoffs.\n\nv3.0.0 | 2026-06-06T07:19:53.599Z | user\n\nA bilingual project memory workflow that records project metadata, targets, current status, completed work, pending work, next actions, archive history, and AI handoff notes at project start, checkpoint, wrap-up, or handoff.\n\n一个中英双语项目记忆工作流 Skill，用于在项目启动、工作中段、收工或交接时保存项目元数据、目标、当前状态、已完成事项、待办事项、下一步计划、归档历史和 AI 交接信息。\n\nUpgrade Introduction / 升级说明\n\nThis is the upgraded version of the previous Chinese and English Daily Workflow editions; going forward, both language editions will be maintained together in this single bilingual package.\n\n这是此前中文与英文 Daily Workflow 两版的升级版；以后中英文版本会合并在这个双语包里统一维护。\n\nv2.0.0 | 2026-05-09T15:19:53.510Z | user\n\nVersion 2.0.0 introduces a project-local, context-compressing, bilingual workflow manager:\n\n- Adds context compression at checkpoint, wrap-up, and handoff, storing only essential info for user and next AI.\n- Unifies start/checkpoint/wrap-up/handoff flows with concise, actionable summaries instead of full transcripts.\n- Modernizes file structure: Docs/ directory with TARGET.md, STATUS.md, COMPLETED.md, PENDING.md, SCHEDULE.md, and optional HANDOFF.md.\n- Supports both Chinese and English trigger phrases and configurable workflow triggers in Docs/CONFIG.md.\n- Strongly focuses on clear AI-to-AI and user handoff, improved privacy, and reduced context bloat.\n- Includes initial project security policy (SECURITY.md) and new agent configuration file.\n\nv1.1.0 | 2026-05-05T16:02:27.789Z | user\n\n### v1.1.0 (2026-05-05)\n✨ 新功能• 新增「吃饭啦」触发点（Phase 1.5）\n— 午休前自动更新四个工作档案• 新增首次运行配置向导 \n— 支持自定义三个触发暗语\n🔧 改进• 配置文件持久化到 .workbuddy/daily-workflow-config.json• 更新工作流图表，完整展示三个触发点流程• 优化文档模板，增加\"上午/下午\"时间段标识\n🐛 修复• 修复收工时不更新 PROJECT_STATUS.md 的问题\n\nv1.0.0 | 2026-05-04T04:23:17.746Z | user\n\n初始版本 1.0.0\n\n功能介绍：\n- ✅ 支持「开工啦」触发词，自动读取项目状态\n- ✅ 支持「收工啦」触发词，自动保存工作进度\n- ✅ 自动管理 Docs/ 目录下的5个核心文档：\n - PROJECT_TARGET.md (项目目标)\n - PROJECT_STATUS.md (项目进度)\n - COMPLETED_JOBS.md (已完成工作)\n - PENDING_JOBS.md (未完成工作)\n - NEXT_STEPS.md (下一步计划)\n- ✅ 自动创建缺失的目录和文档（使用标准模板）\n- ✅ AI-to-AI 无缝交接，保留完整项目上下文\n- ✅ 适用于多 AI 协作的项目管理场景\n\n适用场景：\n- 每日工作开始/结束时的项目状态管理\n- 多人/多 AI 协作时的上下文传递\n- 需要详细工作日志的项目管理\n\nArchive index:\n\nArchive v4.1.0: 6 files, 9620 bytes\n\nFiles: agents/openai.yaml (346b), QUICKSTART.md (1828b), SECURITY.md (2423b), skill-card.md (2068b), SKILL.md (11082b), _meta.json (133b)\n\nFile v4.1.0:SKILL.md\n\n---\nname: daily-workflow\ndescription: Say 开工啦 or 收工啦 and get a resumable project note written for you, so tomorrow you or another AI can pick up without re-reading everything. Use when a user asks to resume a project, save progress, record a checkpoint, end a work session, prepare a handoff, switch to a new AI session, or reconcile stale working notes. Typical triggers include 开工啦, 中段检查, 吃饭啦, 收工啦, 交接, 记一下进度, 存个档, 明天接着做, 换个 AI 继续, 上下文快满了, start work, checkpoint, wrap up, handoff, save progress, resume tomorrow, and switch to a new agent session. Reads existing project authority first, preserves dirty worktrees and governance evidence, avoids competing state files, archives rather than deletes history, and records commands, final outcomes, blockers, risks, and exact next actions without claiming unverified completion or QA acceptance.\n---\n\n# Daily Workflow / 项目记忆工作流\n\nVersion: 4.1.0\n\nUse this skill to make work resumable. Record only the compact factual state needed by the user or next agent; do not turn project memory into a second project-management system.\n\nRespond in the user's language. Keep machine-readable status values in English.\n\n## Scope And Ownership\n\nThis skill owns session continuity only: orientation, checkpoint summaries, wrap-up records, and handoffs.\n\n- `cms-project-governance` owns formal targets, Programs, Work Orders, Controller/QA decisions, and acceptance state.\n- `agent-loop-engineering` owns active coding-loop evidence, stop gates, evaluation, and loop records.\n- `project-lifecycle-navigator` owns lifecycle analysis and rebaseline proposals.\n- `web-search-rules` owns research intake and source rules.\n- `ai-workflow-os` may route here but must not create a parallel memory state.\n\nWhen another system already owns a field, read and summarize it; do not overwrite it.\n\n## Trigger And Authorization\n\nUse this skill when the user explicitly asks to:\n\n- start or resume work;\n- save progress or create a checkpoint;\n- wrap up or end work;\n- prepare a handoff;\n- reconcile or migrate workflow notes.\n\nCommon phrases include `开工啦`, `中段检查`, `吃饭啦`, `收工啦`, `交接`, `start work`, `checkpoint`, `wrap up`, and `handoff`.\n\nA casual mention of one phrase is not enough when intent is ambiguous. Inspect read-only first. Persist files only when the request clearly asks to initialize or update project memory, or when an established project workflow already defines the trigger as a write command.\n\n## Read-Only Orientation First\n\nBefore writing:\n\n1. resolve the actual workspace and repository root, including nested repositories;\n2. inspect Git branch, commit, and dirty state without modifying it;\n3. locate governing files and determine current authority precedence;\n4. locate product scope, entrypoints, build/test commands, and runtime surfaces relevant to the session;\n5. read existing memory files and compatible aliases;\n6. identify contradictions, missing evidence, and ownership conflicts.\n\nDo not use `git reset`, `git clean`, broad deletion, or ambiguous process termination. Existing changes belong to the user unless proven otherwise.\n\n## Choose A Memory Profile\n\n### Existing Governance Profile\n\nUse when project-owned governance or loop files already exist. Follow that project's names and authority. Do not create duplicate `Docs/` files merely because this skill has default templates.\n\n### Lightweight Profile\n\nUse only when no authoritative workflow exists and persistence is requested. Start with the smallest useful set:\n\n```text\nDocs/STATUS.md\nDocs/NEXT_ACTIONS.md\n```\n\nAdd files only when their information has a distinct owner:\n\n```text\nDocs/PROJECT.md project identity and metadata\nDocs/TARGET.md Owner-confirmed target, scope, Non-Goals, success criteria\nDocs/COMPLETED.md append-only historical completion log\nDocs/PENDING.md current queue, blockers, and decisions\nDocs/HANDOFF.md standalone handoff only when requested or too large to embed\nDocs/CONFIG.md explicit local workflow preferences\nDocs/archive/YYYY-MM.md\n```\n\nDo not create `TARGET.md` from an AI guess. If the target is unclear, record a proposed summary in `STATUS.md` as `TBD - Owner Confirmation Required`.\n\n### Legacy Migration Profile\n\nDetect legacy files read-only. If old and new forms coexist, show the conflict and ask before merging. Never delete legacy files automatically.\n\nCompatibility mappings:\n\n```text\nPROJECT_TARGET.md -> TARGET.md\nPROJECT_STATUS.md -> STATUS.md\nCOMPLETED_JOBS.md -> COMPLETED.md\nPENDING_JOBS.md -> PENDING.md\nNEXT_STEPS.md -> NEXT_ACTIONS.md\nSCHEDULE.md -> NEXT_ACTIONS.md compatibility alias\n```\n\n## Evidence Vocabulary\n\nUse precise states:\n\n- `implemented`: source or artifact exists;\n- `verified`: a relevant current check passed;\n- `partial`: only part of the intended behavior exists;\n- `unverified`: required evidence was not run or observed;\n- `unusable`: present but the real user flow cannot complete;\n- `documentation-conflict`: current records disagree;\n- `not-executed`: a scenario was not run;\n- `blocked`: progress requires authority, user input, or an unavailable dependency;\n- `accepted`: only when the authorized independent acceptance role has recorded it.\n\nNever promote Developer self-report, historical logs, a health endpoint, compilation, a narrow test, or a screenshot into broader acceptance.\n\nFor each verification record, preserve:\n\n- exact command or manual scenario;\n- relevant environment or data boundary;\n- final exit, timeout, interruption, or not-executed state;\n- result summary;\n- evidence/artifact path when useful;\n- residual risk.\n\n## Canonical Record Structure\n\n### STATUS.md\n\nKeep a current snapshot plus short history:\n\n```markdown\n# Project Status\n\n## Current State\n- Status:\n- Current goal:\n- Active scope:\n- Latest verified behavior:\n- Blockers:\n- Residual risks:\n\n## Latest Verification\n- Command or scenario:\n- Final result:\n- Evidence boundary:\n- Evidence path:\n- Not executed / deferred:\n\n## Compressed Context\n- User intent:\n- Decisions:\n- Completed since last checkpoint:\n- Pending:\n- Files touched:\n- Dirty worktree notes:\n- Exact next action:\n\n## Update History\n### YYYY-MM-DD HH:mm [start|checkpoint|wrap-up|handoff|migration]\n- Factual change and evidence.\n```\n\n### NEXT_ACTIONS.md\n\nKeep exactly one current continuation plan:\n\n```markdown\n# Next Actions\n\n## Immediate Next Action\n1. [Action with context and acceptance condition]\n\n## Then\n1. [Ordered action]\n\n## Blockers And Owner Decisions\n- [Decision or blocker]\n\n## Handoff Snapshot\n- Current state:\n- Completed:\n- Pending:\n- Risks:\n- Files/artifacts:\n- Commands and final results:\n- Exact next action:\n```\n\n## Start Work\n\n1. Orient read-only.\n2. Determine the memory profile and authority.\n3. Summarize target, current state, latest verified evidence, dirty changes, pending work, blockers, and exact next action.\n4. Reconcile stale notes against current repository/runtime evidence.\n5. Write a start entry only when persistence is authorized.\n6. Do not create speculative targets or mark historical completion as current verification.\n\n## Checkpoint\n\n1. Compress only the context required to continue.\n2. Record actual completed work and current evidence.\n3. Record failed, timed-out, interrupted, deferred, and unrun checks explicitly.\n4. Update the current queue without dropping unresolved blockers or Owner decisions.\n5. Refresh the single exact next action.\n6. Preserve governance and coding-loop fields owned by other skills.\n\n## Wrap Up\n\n1. Record the final state observed in this session.\n2. Append completed history only for work that actually occurred.\n3. Keep current status aligned with the latest verified state; preserve earlier failed attempts in history with timestamps.\n4. List unfinished work, blockers, residual risks, and deferred scenarios.\n5. Do not change the target unless the Owner explicitly changed it.\n6. Prepare a resumable next action.\n\n## Handoff\n\nCreate a self-contained handoff that includes:\n\n- product/architecture context necessary to continue;\n- current authoritative decisions and scope;\n- exact artifacts and files;\n- completed work and what evidence supports it;\n- unverified or not-executed scenarios;\n- blockers, risks, and Owner decisions;\n- commands/tests with final outcomes;\n- exact next actions and prohibited actions.\n\nDo not rely on chat history. Do not include hidden reasoning or a full transcript.\n\n## Atomic Update Rules\n\nWhen several memory files describe one state transition, update them as one coherent change and re-read them afterward. If interrupted, report which files changed and which did not.\n\n- One current fact should have one authoritative home.\n- Append historical records; rewrite current snapshots intentionally.\n- Preserve unresolved blockers.\n- Preserve external ownership fields such as acceptance evidence, QA decisions, loop evaluations, and `LOOP_RUNS.jsonl`.\n- Archive older status entries by moving them to `Docs/archive/YYYY-MM.md`; never delete history automatically.\n- Use UTF-8 for all Markdown and JSONL files.\n- Report absolute paths to the user.\n\n## Memory Bloat Control\n\nSession memory must stay cheap enough to be read at the start of the next session.\n\n- Keep the live memory set small: a current snapshot plus one continuation plan. Everything older moves to `Docs/archive/YYYY-MM.md`.\n- Never annotate a live file with `Superseded`, `CURRENT OVERRIDE`, or `this section is void`. If the target changed, write a new `REBASELINE-.md` and archive the previous one in the same pass. A reader should never have to reconstruct which layer is current.\n- Do not create a new file for information that already has an owner elsewhere. Summarize and link instead.\n- If the memory directory grows while verified behavior does not, report that imbalance as the checkpoint finding rather than adding another note.\n\n## Before A Bulk Restructure\n\nA checkpoint or handoff that precedes a rename, split, migration, or bulk deletion must first confirm a clean commit or an out-of-workspace backup exists. Record the backup location in the checkpoint. Back up first, then change code.\n\n## Safety\n\nNever record API keys, tokens, passwords, cookies, private keys, `.env` values, browser sessions, real personal contact data, counterparty-confidential details, full private customer records, confidential source bodies, large logs, or sensitive exploit detail. A handoff is frequently pasted into a new session, a chat, or an issue tracker, so treat it as a document that travels.\n\nSummarize sensitive context safely and reference only an appropriate non-secret source location.\n\n## User-Facing Summary\n\nKeep the response compact:\n\n```text\nProject memory updated.\n- Current state: Needs Fix\n- Verified: 2 checks passed; browser flow not executed\n- Updated: STATUS.md, NEXT_ACTIONS.md\n- Preserved: active QA and loop evidence\n- Blocker: Owner decision on target scope\n- Next: reproduce the failing user-visible flow with the current data boundary\n```\n\nFile v4.1.0:_meta.json\n\n{\n \"ownerId\": \"kn7bjt2sd8f7cq83y0nf33xt19855gv6\",\n \"slug\": \"daily-workflow\",\n \"version\": \"4.1.0\",\n \"publishedAt\": 1788800774694\n}\n\nFile v4.1.0:QUICKSTART.md\n\n# Quickstart / 快速开始\n\n## Install / 安装\n\n```bash\nopenclaw skills install daily-workflow\n```\n\nor / 或\n\n```bash\nclawhub install daily-workflow\n```\n\n## Try it in 30 seconds / 30 秒验证\n\nSay this to your agent at the end of a work session / 收工时对 agent 说：\n\n```text\n收工啦\n```\n\nYou should get / 你应该得到：\n\n- `Docs/STATUS.md` — current state, latest verification, next action;\n- `Docs/NEXT_ACTIONS.md` — the immediate next step, then what follows,\n blockers and Owner decisions;\n- a one-line summary of what was written.\n\nOpen those two files. That is the entire output — no dashboard, no database.\n\n## Four phrases are enough / 四句话就够\n\n| Say | What happens |\n| --- | --- |\n| `开工啦` | read existing state, orient for the session |\n| `中段检查` | mid-session checkpoint |\n| `收工啦` | wrap up, record exact next actions |\n| `交接` | package a handoff for another person or another AI |\n\nThe point is resumability: tomorrow you, or a fresh AI session, can pick up\nwithout re-reading the whole project.\n\n## Minimum useful path / 最小可用路径\n\nJust use `开工啦` and `收工啦`.\n\nYou do **not** need `Docs/PROJECT.md`, `TARGET.md`, `COMPLETED.md` or\n`PENDING.md`. Those belong to the governance profile and only appear if your\nproject already runs one. The default lightweight profile writes two files.\n\n## Privacy and safety / 隐私与安全\n\n- Runs locally. Writes plain Markdown into your project's own `Docs/`.\n- Stores no API keys, tokens, passwords or credentials.\n- Never deletes history — superseded entries are archived, not removed.\n- Read-only orientation first: nothing is written until intent is clear.\n- Will not claim work is complete or QA-accepted without evidence.\n\n## Next / 下一步\n\nFull record structure and memory profiles: `SKILL.md`.\n\nFile v4.1.0:SECURITY.md\n\n# Daily Workflow / 项目记忆工作流 - Security Guide\n\nVersion: 4.1.0\n\n## Security Statement / 安全说明\n\nThis skill may write project-local workflow notes when the user explicitly requests persistence or an established trigger authorizes it. It is designed for resumability and handoff, not for storing secrets, replacing project governance, or recording confidential source bodies.\n\n本 Skill 只应在项目本地 `Docs/` 中写入工作流笔记，用于恢复上下文和交接，不用于保存密钥、账号凭证或完整敏感资料。\n\n## Allowed Storage / 允许写入范围\n\nDefault allowed write scope when no project-owned workflow defines a narrower location:\n\n```text\nDocs/\n```\n\nDo not write global user state. Do not write outside `Docs/` unless the user explicitly asks as part of the broader task.\n\nBefore writing, resolve the actual workspace and repository, preserve dirty changes, identify existing governance ownership, and avoid creating competing state files.\n\n## Sensitive Data Rules / 敏感信息规则\n\nNever record:\n\n- API keys, tokens, passwords, private keys, cookies, OAuth refresh tokens, browser sessions.\n- `.env` values or credential files.\n- Full private customer records.\n- Long confidential source excerpts.\n- Large unrelated logs.\n\nIf sensitive context is relevant, record a safe summary and the non-secret location where the user can find the source.\n\n## Legacy Migration Safety / 旧版迁移安全\n\nLegacy files may be read for migration, but must not be deleted automatically. If both old and new files exist, ask before merging. Do not migrate secret-like fields.\n\n## Archive Safety / 归档安全\n\nArchiving older status history is allowed only by moving content into `Docs/archive/`. Never delete history automatically.\n\n## Release Checklist / 发布检查\n\n- `SKILL.md` and `SECURITY.md` show version `4.1.0`; generated registry snapshots are excluded from the publish bundle.\n- The package uses one bilingual skill, not separate English and Chinese packages.\n- Legacy file migration is documented.\n- `NEXT_ACTIONS.md` is the primary continuation file, with `SCHEDULE.md` treated as compatibility alias.\n- Existing project/governance files remain authoritative; the lightweight profile starts with only `STATUS.md` and `NEXT_ACTIONS.md`.\n- Completion, verification, and independent acceptance remain separate.\n- No example contains real credentials or private data.\n\nFile v4.1.0:skill-card.md\n\n## Description:\n\nDaily Workflow helps agents create evidence-backed project checkpoints, wrap-ups, and handoffs so work can resume without re-reading the whole project.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[englandtong](https://clawhub.ai/user/englandtong)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and agent users use this skill to preserve concise project memory at start, checkpoint, wrap-up, and handoff moments. It writes project-local Markdown notes that capture current state, verification evidence, blockers, risks, and exact next actions without replacing existing governance records.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Checkpoint or handoff notes can contain sensitive project context if shared outside the project.\n\nMitigation: Review generated Docs/ notes before sharing and avoid recording secrets, credentials, full confidential data, or large private logs.\n\nRisk: Project memory files can conflict with existing governance or acceptance records.\n\nMitigation: Read existing project authority first, preserve dirty worktrees and governance evidence, and summarize externally owned fields instead of overwriting them.\n\n## Reference(s):\n\n- [ClawHub Skill Page](https://clawhub.ai/englandtong/skills/daily-workflow)\n- [Quickstart](artifact/QUICKSTART.md)\n- [Security Guide](artifact/SECURITY.md)\n\n## Skill Output:\n\n**Output Type(s):** [Markdown, Files, Guidance]\n\n**Output Format:** [Markdown files and concise user-facing summaries]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Writes project-local workflow notes, normally under Docs/, when persistence is clearly authorized.]\n\n## Skill Version(s):\n\n4.1.0 (source: server release metadata and artifact SKILL.md)\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\nFile v4.1.0:agents/openai.yaml\n\ninterface:\n display_name: \"Daily Workflow\"\n short_description: \"Preserve evidence-backed project memory\"\n default_prompt: \"Use $daily-workflow to orient from the current project state and create an evidence-backed checkpoint, wrap-up, or handoff without overwriting governance or acceptance records.\"\npolicy:\n allow_implicit_invocation: true\n\nArchive v4.0.0: 5 files, 7607 bytes\n\nFiles: agents/openai.yaml (346b), SECURITY.md (2423b), skill-card.md (2003b), SKILL.md (9522b), _meta.json (133b)\n\nFile v4.0.0:SKILL.md\n\n---\nname: daily-workflow\ndescription: Preserve concise, evidence-backed project memory across start-work orientation, checkpoints, wrap-up, and handoff. Use when a user explicitly asks to resume a project, save progress, record a checkpoint, end a work session, prepare a self-contained handoff, or reconcile stale working notes. Reads existing project authority first, preserves dirty worktrees and governance evidence, avoids competing state files, and records commands, final outcomes, blockers, risks, and exact next actions without claiming unverified completion or QA acceptance.\n---\n\n# Daily Workflow / 项目记忆工作流\n\nVersion: 4.0.0\n\nUse this skill to make work resumable. Record only the compact factual state needed by the user or next agent; do not turn project memory into a second project-management system.\n\nRespond in the user's language. Keep machine-readable status values in English.\n\n## Scope And Ownership\n\nThis skill owns session continuity only: orientation, checkpoint summaries, wrap-up records, and handoffs.\n\n- `cms-project-governance` owns formal targets, Programs, Work Orders, Controller/QA decisions, and acceptance state.\n- `agent-loop-engineering` owns active coding-loop evidence, stop gates, evaluation, and loop records.\n- `project-lifecycle-navigator` owns lifecycle analysis and rebaseline proposals.\n- `web-search-rules` owns research intake and source rules.\n- `ai-workflow-os` may route here but must not create a parallel memory state.\n\nWhen another system already owns a field, read and summarize it; do not overwrite it.\n\n## Trigger And Authorization\n\nUse this skill when the user explicitly asks to:\n\n- start or resume work;\n- save progress or create a checkpoint;\n- wrap up or end work;\n- prepare a handoff;\n- reconcile or migrate workflow notes.\n\nCommon phrases include `开工啦`, `中段检查`, `吃饭啦`, `收工啦`, `交接`, `start work`, `checkpoint`, `wrap up`, and `handoff`.\n\nA casual mention of one phrase is not enough when intent is ambiguous. Inspect read-only first. Persist files only when the request clearly asks to initialize or update project memory, or when an established project workflow already defines the trigger as a write command.\n\n## Read-Only Orientation First\n\nBefore writing:\n\n1. resolve the actual workspace and repository root, including nested repositories;\n2. inspect Git branch, commit, and dirty state without modifying it;\n3. locate governing files and determine current authority precedence;\n4. locate product scope, entrypoints, build/test commands, and runtime surfaces relevant to the session;\n5. read existing memory files and compatible aliases;\n6. identify contradictions, missing evidence, and ownership conflicts.\n\nDo not use `git reset`, `git clean`, broad deletion, or ambiguous process termination. Existing changes belong to the user unless proven otherwise.\n\n## Choose A Memory Profile\n\n### Existing Governance Profile\n\nUse when project-owned governance or loop files already exist. Follow that project's names and authority. Do not create duplicate `Docs/` files merely because this skill has default templates.\n\n### Lightweight Profile\n\nUse only when no authoritative workflow exists and persistence is requested. Start with the smallest useful set:\n\n```text\nDocs/STATUS.md\nDocs/NEXT_ACTIONS.md\n```\n\nAdd files only when their information has a distinct owner:\n\n```text\nDocs/PROJECT.md project identity and metadata\nDocs/TARGET.md Owner-confirmed target, scope, Non-Goals, success criteria\nDocs/COMPLETED.md append-only historical completion log\nDocs/PENDING.md current queue, blockers, and decisions\nDocs/HANDOFF.md standalone handoff only when requested or too large to embed\nDocs/CONFIG.md explicit local workflow preferences\nDocs/archive/YYYY-MM.md\n```\n\nDo not create `TARGET.md` from an AI guess. If the target is unclear, record a proposed summary in `STATUS.md` as `TBD - Owner Confirmation Required`.\n\n### Legacy Migration Profile\n\nDetect legacy files read-only. If old and new forms coexist, show the conflict and ask before merging. Never delete legacy files automatically.\n\nCompatibility mappings:\n\n```text\nPROJECT_TARGET.md -> TARGET.md\nPROJECT_STATUS.md -> STATUS.md\nCOMPLETED_JOBS.md -> COMPLETED.md\nPENDING_JOBS.md -> PENDING.md\nNEXT_STEPS.md -> NEXT_ACTIONS.md\nSCHEDULE.md -> NEXT_ACTIONS.md compatibility alias\n```\n\n## Evidence Vocabulary\n\nUse precise states:\n\n- `implemented`: source or artifact exists;\n- `verified`: a relevant current check passed;\n- `partial`: only part of the intended behavior exists;\n- `unverified`: required evidence was not run or observed;\n- `unusable`: present but the real user flow cannot complete;\n- `documentation-conflict`: current records disagree;\n- `not-executed`: a scenario was not run;\n- `blocked`: progress requires authority, user input, or an unavailable dependency;\n- `accepted`: only when the authorized independent acceptance role has recorded it.\n\nNever promote Developer self-report, historical logs, a health endpoint, compilation, a narrow test, or a screenshot into broader acceptance.\n\nFor each verification record, preserve:\n\n- exact command or manual scenario;\n- relevant environment or data boundary;\n- final exit, timeout, interruption, or not-executed state;\n- result summary;\n- evidence/artifact path when useful;\n- residual risk.\n\n## Canonical Record Structure\n\n### STATUS.md\n\nKeep a current snapshot plus short history:\n\n```markdown\n# Project Status\n\n## Current State\n- Status:\n- Current goal:\n- Active scope:\n- Latest verified behavior:\n- Blockers:\n- Residual risks:\n\n## Latest Verification\n- Command or scenario:\n- Final result:\n- Evidence boundary:\n- Evidence path:\n- Not executed / deferred:\n\n## Compressed Context\n- User intent:\n- Decisions:\n- Completed since last checkpoint:\n- Pending:\n- Files touched:\n- Dirty worktree notes:\n- Exact next action:\n\n## Update History\n### YYYY-MM-DD HH:mm [start|checkpoint|wrap-up|handoff|migration]\n- Factual change and evidence.\n```\n\n### NEXT_ACTIONS.md\n\nKeep exactly one current continuation plan:\n\n```markdown\n# Next Actions\n\n## Immediate Next Action\n1. [Action with context and acceptance condition]\n\n## Then\n1. [Ordered action]\n\n## Blockers And Owner Decisions\n- [Decision or blocker]\n\n## Handoff Snapshot\n- Current state:\n- Completed:\n- Pending:\n- Risks:\n- Files/artifacts:\n- Commands and final results:\n- Exact next action:\n```\n\n## Start Work\n\n1. Orient read-only.\n2. Determine the memory profile and authority.\n3. Summarize target, current state, latest verified evidence, dirty changes, pending work, blockers, and exact next action.\n4. Reconcile stale notes against current repository/runtime evidence.\n5. Write a start entry only when persistence is authorized.\n6. Do not create speculative targets or mark historical completion as current verification.\n\n## Checkpoint\n\n1. Compress only the context required to continue.\n2. Record actual completed work and current evidence.\n3. Record failed, timed-out, interrupted, deferred, and unrun checks explicitly.\n4. Update the current queue without dropping unresolved blockers or Owner decisions.\n5. Refresh the single exact next action.\n6. Preserve governance and coding-loop fields owned by other skills.\n\n## Wrap Up\n\n1. Record the final state observed in this session.\n2. Append completed history only for work that actually occurred.\n3. Keep current status aligned with the latest verified state; preserve earlier failed attempts in history with timestamps.\n4. List unfinished work, blockers, residual risks, and deferred scenarios.\n5. Do not change the target unless the Owner explicitly changed it.\n6. Prepare a resumable next action.\n\n## Handoff\n\nCreate a self-contained handoff that includes:\n\n- product/architecture context necessary to continue;\n- current authoritative decisions and scope;\n- exact artifacts and files;\n- completed work and what evidence supports it;\n- unverified or not-executed scenarios;\n- blockers, risks, and Owner decisions;\n- commands/tests with final outcomes;\n- exact next actions and prohibited actions.\n\nDo not rely on chat history. Do not include hidden reasoning or a full transcript.\n\n## Atomic Update Rules\n\nWhen several memory files describe one state transition, update them as one coherent change and re-read them afterward. If interrupted, report which files changed and which did not.\n\n- One current fact should have one authoritative home.\n- Append historical records; rewrite current snapshots intentionally.\n- Preserve unresolved blockers.\n- Preserve external ownership fields such as acceptance evidence, QA decisions, loop evaluations, and `LOOP_RUNS.jsonl`.\n- Archive older status entries by moving them to `Docs/archive/YYYY-MM.md`; never delete history automatically.\n- Use UTF-8 for all Markdown and JSONL files.\n- Report absolute paths to the user.\n\n## Safety\n\nNever record API keys, tokens, passwords, cookies, private keys, `.env` values, browser sessions, full private customer records, confidential source bodies, large logs, or sensitive exploit detail.\n\nSummarize sensitive context safely and reference only an appropriate non-secret source location.\n\n## User-Facing Summary\n\nKeep the response compact:\n\n```text\nProject memory updated.\n- Current state: Needs Fix\n- Verified: 2 checks passed; browser flow not executed\n- Updated: STATUS.md, NEXT_ACTIONS.md\n- Preserved: active QA and loop evidence\n- Blocker: Owner decision on target scope\n- Next: reproduce the failing user-visible flow with the current data boundary\n```\n\nFile v4.0.0:_meta.json\n\n{\n \"ownerId\": \"kn7bjt2sd8f7cq83y0nf33xt19855gv6\",\n \"slug\": \"daily-workflow\",\n \"version\": \"4.0.0\",\n \"publishedAt\": 1786719776272\n}\n\nFile v4.0.0:SECURITY.md\n\n# Daily Workflow / 项目记忆工作流 - Security Guide\n\nVersion: 4.0.0\n\n## Security Statement / 安全说明\n\nThis skill may write project-local workflow notes when the user explicitly requests persistence or an established trigger authorizes it. It is designed for resumability and handoff, not for storing secrets, replacing project governance, or recording confidential source bodies.\n\n本 Skill 只应在项目本地 `Docs/` 中写入工作流笔记，用于恢复上下文和交接，不用于保存密钥、账号凭证或完整敏感资料。\n\n## Allowed Storage / 允许写入范围\n\nDefault allowed write scope when no project-owned workflow defines a narrower location:\n\n```text\nDocs/\n```\n\nDo not write global user state. Do not write outside `Docs/` unless the user explicitly asks as part of the broader task.\n\nBefore writing, resolve the actual workspace and repository, preserve dirty changes, identify existing governance ownership, and avoid creating competing state files.\n\n## Sensitive Data Rules / 敏感信息规则\n\nNever record:\n\n- API keys, tokens, passwords, private keys, cookies, OAuth refresh tokens, browser sessions.\n- `.env` values or credential files.\n- Full private customer records.\n- Long confidential source excerpts.\n- Large unrelated logs.\n\nIf sensitive context is relevant, record a safe summary and the non-secret location where the user can find the source.\n\n## Legacy Migration Safety / 旧版迁移安全\n\nLegacy files may be read for migration, but must not be deleted automatically. If both old and new files exist, ask before merging. Do not migrate secret-like fields.\n\n## Archive Safety / 归档安全\n\nArchiving older status history is allowed only by moving content into `Docs/archive/`. Never delete history automatically.\n\n## Release Checklist / 发布检查\n\n- `SKILL.md` and `SECURITY.md` show version `4.0.0`; generated registry snapshots are excluded from the publish bundle.\n- The package uses one bilingual skill, not separate English and Chinese packages.\n- Legacy file migration is documented.\n- `NEXT_ACTIONS.md` is the primary continuation file, with `SCHEDULE.md` treated as compatibility alias.\n- Existing project/governance files remain authoritative; the lightweight profile starts with only `STATUS.md` and `NEXT_ACTIONS.md`.\n- Completion, verification, and independent acceptance remain separate.\n- No example contains real credentials or private data.\n\nFile v4.0.0:skill-card.md\n\n## Description:\n\nDaily Workflow preserves concise, evidence-backed project memory across start-work orientation, checkpoints, wrap-up, and handoff without overwriting governance or acceptance records.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[englandtong](https://clawhub.ai/user/englandtong)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and project teams use this skill to resume work, save progress, record checkpoints, wrap up sessions, prepare handoffs, and reconcile stale workflow notes while preserving project governance boundaries.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Workflow notes may accidentally retain sensitive project context if users ask the agent to persist too much detail.\n\nMitigation: Review generated notes before sharing and avoid recording secrets, credentials, private customer records, large logs, or confidential source bodies.\n\nRisk: Project memory can become misleading if it overwrites governance records or treats unverified work as accepted.\n\nMitigation: Use read-only orientation first, preserve existing authority, keep verification and acceptance separate, and record blockers or unexecuted checks explicitly.\n\n## Reference(s):\n\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, shell commands, configuration, guidance]\n\n**Output Format:** [Markdown guidance and project-local workflow note updates]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May create or update scoped project memory files such as Docs/STATUS.md, Docs/NEXT_ACTIONS.md, or handoff notes when persistence is authorized.]\n\n## Skill Version(s):\n\n4.0.0 (source: server release metadata, SKILL.md, SECURITY.md)\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\nFile v4.0.0:agents/openai.yaml\n\ninterface:\n display_name: \"Daily Workflow\"\n short_description: \"Preserve evidence-backed project memory\"\n default_prompt: \"Use $daily-workflow to orient from the current project state and create an evidence-backed checkpoint, wrap-up, or handoff without overwriting governance or acceptance records.\"\npolicy:\n allow_implicit_invocation: true\n\nArchive v3.0.0: 6 files, 10004 bytes\n\nFiles: _meta.json (133b), agents/openai.yaml (508b), SECURITY.md (1833b), sitemap.xml (3551b), skill-card.md (2356b), SKILL.md (12917b)\n\nFile v3.0.0:SKILL.md\n\n---\nname: daily-workflow\ndescription: \"Bilingual EN/ZH project memory workflow for start work, checkpoint, wrap-up, and handoff. Maintains project-local Docs/ notes with target, metadata, status, compressed context, completed work, pending work, next actions, archives, and legacy migration from older Daily Workflow file names.\"\n---\n\n# Daily Workflow / 项目记忆工作流\n\nVersion: 3.0.0 \nPurpose: keep a project understandable and resumable across work sessions, context resets, and AI-to-AI handoffs. \nStorage: project-local `Docs/` only.\nUpgrade note: This is the upgraded version of the previous Chinese and English Daily Workflow editions; going forward, both language editions will be maintained together in this single bilingual package. / 这是此前中文与英文 Daily Workflow 两版的升级版；以后中英文版本会合并在这个双语包里统一维护。\n\n## Core Principle / 核心原则\n\nWrite concise, factual project state that a user and the next AI can continue from. Prefer concrete targets, decisions, files, commands, blockers, risks, and next actions over narrative.\n\n记录应当简洁、可验证、可继续执行。优先保存目标、决策、文件路径、命令结果、阻塞、风险和下一步，而不是聊天记录全文。\n\nDo not record secrets, credentials, private customer data, full confidential documents, large logs, or unrelated source excerpts.\n\n## Trigger Phrases / 触发语\n\nTreat these phrases as equivalent triggers:\n\n- Start work / 开工: `开工啦`, `开始工作`, `start work`, `begin work`, `starting work`, `start of day`\n- Checkpoint / 中段检查: `中段检查`, `吃饭啦`, `checkpoint`, `lunch check`, `lunch time`, `break time`, `midday check`\n- Wrap up / 收工: `收工啦`, `结束工作`, `wrap up`, `end work`, `ending work`, `end of day`\n- Handoff / 交接: `交接`, `handoff`, `next AI`, `下一任 AI`\n- Reconfigure / 重新配置: `重新配置工作流`, `configure workflow`, `update workflow config`\n\nIf the user defines custom triggers in `Docs/CONFIG.md`, honor them.\n\n## Managed Files / 管理文件\n\nCreate `Docs/` if missing. Maintain these files:\n\n- `Docs/PROJECT.md`: project metadata, owner, workspace, repository, phase, language, and last checkpoint.\n- `Docs/TARGET.md`: project purpose, success criteria, scope, and out-of-scope items. Default read-only after creation unless the user changes the goal.\n- `Docs/STATUS.md`: current state, progress, active context, compressed context, latest update history, and known risks.\n- `Docs/COMPLETED.md`: completed work log. Append only; never overwrite older entries.\n- `Docs/PENDING.md`: current pending work, blockers, and decisions needed. May be rewritten as the current queue while preserving unresolved blockers.\n- `Docs/NEXT_ACTIONS.md`: immediate continuation plan and handoff plan. Overwrite each checkpoint/wrap-up with the latest actionable plan.\n- `Docs/CONFIG.md`: optional trigger and language preferences.\n\nCompatibility alias:\n\n- `Docs/SCHEDULE.md` is a legacy/current alias for next actions. If it exists, read it. Prefer writing `NEXT_ACTIONS.md` in v3.0.0. If a project already depends on `SCHEDULE.md`, either keep both synchronized or explain the migration before changing.\n\nOptional files:\n\n- `Docs/HANDOFF.md`: create only when the user asks for a standalone handoff file or when the handoff block becomes too long.\n- `Docs/archive/YYYY-MM.md`: monthly archive for older status updates.\n\n## Configuration / 配置\n\nOptional configuration is project-local:\n\n```text\nDocs/CONFIG.md\n```\n\nDefault config block:\n\n```json\n{\n \"version\": \"3.0.0\",\n \"startPhrase\": \"开工啦\",\n \"checkpointPhrase\": \"中段检查\",\n \"lunchPhrase\": \"吃饭啦\",\n \"endPhrase\": \"收工啦\",\n \"handoffPhrase\": \"交接\",\n \"language\": \"bilingual\",\n \"primaryNextActionsFile\": \"NEXT_ACTIONS.md\",\n \"archiveAfterStatusEntries\": 10,\n \"firstRun\": false\n}\n```\n\nCreate or update config only inside `Docs/`. Do not write global user state.\n\n## Legacy Migration / 旧版迁移\n\nBefore creating new files, check for legacy Daily Workflow files.\n\nLegacy English v1 names:\n\n```text\nDocs/PROJECT_TARGET.md -> Docs/TARGET.md\nDocs/PROJECT_STATUS.md -> Docs/STATUS.md\nDocs/COMPLETED_JOBS.md -> Docs/COMPLETED.md\nDocs/PENDING_JOBS.md -> Docs/PENDING.md\nDocs/NEXT_STEPS.md -> Docs/NEXT_ACTIONS.md\n.workbuddy/daily-workflow-config.json -> Docs/CONFIG.md\n```\n\nLegacy v2/current alias:\n\n```text\nDocs/SCHEDULE.md -> Docs/NEXT_ACTIONS.md\n```\n\nMigration rules:\n\n1. Read legacy files before creating replacements.\n2. If the v3 target file is missing, create it from the legacy content and add `Migrated from: `.\n3. If both legacy and v3 files exist, do not overwrite. Summarize the difference and ask the user whether to merge.\n4. Never delete legacy files automatically.\n5. Never migrate secrets or secret-like fields.\n6. Add a migration entry to `STATUS.md` or `Docs/archive/YYYY-MM.md`.\n\n## File Templates / 文件模板\n\n### PROJECT.md\n\n```markdown\n# Project Metadata / 项目元数据\n\n- Project ID / 项目编号:\n- Project Name / 项目名称:\n- Owner / 负责人:\n- Workspace / 工作区:\n- Repository / 代码仓库:\n- Current Phase / 当前阶段:\n- Primary Language / 主要语言:\n- Risk Level / 风险等级:\n- Last Checkpoint / 最近检查点:\n- Last Updated / 最近更新:\n```\n\n### TARGET.md\n\n```markdown\n# Project Target / 项目目标\n\n## Purpose / 目的\n[What the project is trying to accomplish.]\n\n## Success Criteria / 成功标准\n- [Criterion 1]\n- [Criterion 2]\n\n## Scope / 范围\nIn scope / 范围内:\n- [Item]\n\nOut of scope / 范围外:\n- [Item]\n\n## Assumptions / 假设\n- [Assumption]\n\n## Last Reviewed / 最后确认\n[YYYY-MM-DD]\n```\n\nIf the target is unclear, create a provisional section instead of stopping:\n\n```markdown\n## Unconfirmed Target / 待确认目标\n[Best-effort summary based on current context. Ask the user to confirm or correct.]\n```\n\n### STATUS.md\n\n```markdown\n# Project Status / 项目状态\n\n## Current State / 当前状态\n[Short factual summary.]\n\n## Progress / 进度\n[Percent or milestone-based progress when known.]\n\n## Active Context / 当前上下文\n- Files or areas being worked on:\n- Important constraints:\n- Known risks:\n\n## Compressed Context / 提炼上下文\n- User intent / 用户意图:\n- Decisions made / 已作决策:\n- Current state / 当前状态:\n- Completed work / 已完成:\n- Pending work / 待处理:\n- Blockers / risks / 阻塞与风险:\n- Files touched / 涉及文件:\n- Commands/tests run / 已运行命令或测试:\n- Immediate next action / 下一步立即行动:\n\n## Update History / 更新历史\n### [YYYY-MM-DD HH:mm] [start/checkpoint/wrap-up/handoff/migration]\n- [What changed]\n- [Evidence, command, or file reference when useful]\n```\n\n### COMPLETED.md\n\n```markdown\n# Completed Jobs / 已完成工作\n\n## [YYYY-MM-DD]\n- [Completed item with enough detail for another AI to understand what changed.]\n```\n\n### PENDING.md\n\n```markdown\n# Pending Jobs / 待办工作\n\n## Immediate / 立即处理\n- [ ] [Task, owner/context if known, acceptance condition]\n\n## Later / 稍后处理\n- [ ] [Task]\n\n## Blockers and Decisions / 阻塞与待决策\n- [Blocker or decision needed]\n```\n\n### NEXT_ACTIONS.md\n\n```markdown\n# Next Actions / 下一步行动\n\n## Immediate Next Action / 下一步立即行动\n1. [Action with file/path/context]\n2. [Action]\n\n## Handoff Summary / 交接摘要\n- Current state / 当前状态:\n- Completed since last checkpoint / 上次检查后完成:\n- Pending work / 待处理:\n- Blockers / risks / 阻塞与风险:\n- Files touched / 涉及文件:\n- Commands/tests run / 已运行命令或测试:\n- Immediate next action / 下一步立即行动:\n\n## Notes for Next AI / 给下一任 AI 的说明\n[Any context the next AI must know before acting.]\n```\n\n## Context Compression and Reset / 上下文提炼与清理\n\nAt checkpoint, wrap-up, and handoff time, compress the current conversation context before updating workflow files.\n\nCompression means extracting only what is needed to continue work:\n\n- Current user intent and target\n- Decisions already made\n- Completed work\n- Pending work\n- Blockers, risks, and assumptions\n- Files touched or important paths\n- Commands/tests run and outcomes\n- Immediate next action\n\nWrite this into `Docs/STATUS.md` under the latest update entry and into `Docs/NEXT_ACTIONS.md` for continuation. Do not paste the full conversation transcript.\n\nAfter saving compressed context, treat the old long working context as cleared for workflow purposes. Continue from `STATUS.md` and `NEXT_ACTIONS.md`.\n\nFor checkpoint only, return the compressed context to the current AI in the response so work can continue immediately.\n\n## Start Work Flow / 开工流程\n\nWhen triggered by start work:\n\n1. Check `Docs/`; create missing managed files from templates.\n2. Detect and offer legacy migration when legacy files exist.\n3. Check `Docs/CONFIG.md`; honor it if present.\n4. Read all managed files and compatible legacy aliases.\n5. Summarize target, metadata, current status, recently completed work, pending work, blockers, and immediate next action.\n6. If target is missing or unclear, create or preserve an `Unconfirmed Target` section and ask the user to confirm or correct it.\n7. Update `STATUS.md` with a start entry.\n8. Refresh `NEXT_ACTIONS.md` only if next steps are missing, stale, or contradicted by the user.\n\n## Checkpoint Flow / 中段检查流程\n\nWhen triggered by checkpoint or lunch check:\n\n1. Compress the current conversation into a concise continuation summary.\n2. Append completed items to `COMPLETED.md`.\n3. Update `PROJECT.md` last checkpoint and phase if known.\n4. Update `STATUS.md` with current state, progress, risks, commands/tests run, and compressed context.\n5. Rewrite `PENDING.md` as the current queue while preserving unresolved blockers.\n6. Rewrite `NEXT_ACTIONS.md` with the latest continuation and handoff block.\n7. Archive older `STATUS.md` entries if the configured threshold is exceeded.\n8. Return the compressed context plus the immediate next action.\n\n## Wrap Up Flow / 收工流程\n\nWhen triggered by wrap up:\n\n1. Compress the current conversation into an end-of-session summary.\n2. Append all completed work to `COMPLETED.md`.\n3. Update `PROJECT.md` last checkpoint and phase if known.\n4. Update `STATUS.md` with final state, commands/tests run, risks, and compressed context.\n5. Update `PENDING.md` with unfinished work, blockers, and decisions needed.\n6. Update `TARGET.md` only if the goal changed or the user explicitly asks.\n7. Rewrite `NEXT_ACTIONS.md` as the main handoff document.\n8. Archive older `STATUS.md` entries if needed.\n9. Give the user a short wrap-up summary and the immediate next action for the next session.\n\n## Handoff Flow / 交接流程\n\nWhen triggered by handoff:\n\n1. Read the managed files.\n2. Compress the current conversation into a concise handoff summary.\n3. Update `PROJECT.md`, `STATUS.md`, and `NEXT_ACTIONS.md`.\n4. Include only actionable context: current state, completed work since last checkpoint, pending tasks, blockers and risks, touched files, commands/tests run and outcomes, and immediate next action.\n5. Treat previous long context as cleared for workflow purposes.\n6. Do not invent results that were not observed.\n7. Create `Docs/HANDOFF.md` only if the user asks for a separate handoff artifact.\n\n## Archive Policy / 归档规则\n\nKeep `STATUS.md` readable.\n\n- Keep the latest 5-10 update entries in `STATUS.md`, using `archiveAfterStatusEntries` from config when available.\n- Move older update entries to `Docs/archive/YYYY-MM.md`.\n- Never delete history.\n- Add archive notes with source file, date range, and reason.\n\n## Write Rules / 写入规则\n\n- Use absolute paths when reporting file locations to the user.\n- Append to `COMPLETED.md`; do not overwrite it.\n- Preserve `STATUS.md` update history unless archiving older entries.\n- Preserve unresolved blockers when rewriting `PENDING.md`.\n- Overwrite `NEXT_ACTIONS.md` at each checkpoint or wrap-up so the next AI has a single current plan.\n- Never delete managed or legacy files automatically.\n- Do not edit files outside `Docs/` for this workflow unless the user's broader task requires it.\n\n## Safety Rules / 安全规则\n\nDo not write:\n\n- API keys, tokens, passwords, cookies, private keys, `.env` values, or browser sessions.\n- Full private customer records or confidential source text.\n- Sensitive security findings beyond concise status summaries.\n- Large unrelated logs.\n\nIf important context contains sensitive details, summarize safely and refer to the secure source location only when appropriate.\n\n## User Response Style / 用户反馈格式\n\nKeep responses short and useful:\n\n```text\nSaved workflow state.\n- Updated: PROJECT.md, STATUS.md, NEXT_ACTIONS.md\n- Appended: COMPLETED.md\n- Pending: 3 items, 1 blocker\n- Next: run the failing test and inspect the parser change\n```\n\nFile v3.0.0:_meta.json\n\n{\n \"ownerId\": \"kn7bjt2sd8f7cq83y0nf33xt19855gv6\",\n \"slug\": \"daily-workflow\",\n \"version\": \"3.0.0\",\n \"publishedAt\": 1780730393599\n}\n\nFile v3.0.0:SECURITY.md\n\n# Daily Workflow / 项目记忆工作流 - Security Guide\n\nVersion: 3.0.0\n\n## Security Statement / 安全说明\n\nThis skill writes project-local workflow notes under `Docs/`. It is designed for resumability and handoff, not for storing secrets or confidential records.\n\n本 Skill 只应在项目本地 `Docs/` 中写入工作流笔记，用于恢复上下文和交接，不用于保存密钥、账号凭证或完整敏感资料。\n\n## Allowed Storage / 允许写入范围\n\nDefault allowed write scope:\n\n```text\nDocs/\n```\n\nDo not write global user state. Do not write outside `Docs/` unless the user explicitly asks as part of the broader task.\n\n## Sensitive Data Rules / 敏感信息规则\n\nNever record:\n\n- API keys, tokens, passwords, private keys, cookies, OAuth refresh tokens, browser sessions.\n- `.env` values or credential files.\n- Full private customer records.\n- Long confidential source excerpts.\n- Large unrelated logs.\n\nIf sensitive context is relevant, record a safe summary and the non-secret location where the user can find the source.\n\n## Legacy Migration Safety / 旧版迁移安全\n\nLegacy files may be read for migration, but must not be deleted automatically. If both old and new files exist, ask before merging. Do not migrate secret-like fields.\n\n## Archive Safety / 归档安全\n\nArchiving older status history is allowed only by moving content into `Docs/archive/`. Never delete history automatically.\n\n## Release Checklist / 发布检查\n\n- `SKILL.md`, `SECURITY.md`, `_meta.json`, and `skill-card.md` show version `3.0.0`.\n- The package uses one bilingual skill, not separate English and Chinese packages.\n- Legacy file migration is documented.\n- `NEXT_ACTIONS.md` is the primary continuation file, with `SCHEDULE.md` treated as compatibility alias.\n- No example contains real credentials or private data.\n\nFile v3.0.0:skill-card.md\n\n## Description:
\nBilingual EN/ZH project memory workflow for start work, checkpoint, wrap-up, and handoff. Maintains project-local Docs/ notes with target, metadata, status, compressed context, completed work, pending work, next actions, archives, and legacy migration from older Daily Workflow file names.
\n\nThis skill is ready for commercial/non-commercial use.
\n\n## Publisher:
\n[englandtong](https://clawhub.ai/user/englandtong)
\n\n### License/Terms of Use:
\nMIT-0
\n\n\n## Use Case:
\nDevelopers and project teams use this skill to keep project-local workflow memory current at start work, checkpoints, wrap-up, and handoff. It records concise project metadata, status, completed work, pending work, next actions, archives, and AI handoff notes under Docs/ so work can resume with clear context.
\n\n### Deployment Geography for Use:
\nGlobal
\n\n## Known Risks and Mitigations:
\nRisk: The skill maintains project-local Docs/ workflow files and may record sensitive project notes if the workspace contains private context.
\nMitigation: Review changes to Docs/ before sharing or committing them, and keep secrets, credentials, private customer data, full confidential documents, and large logs out of workflow notes.
\nRisk: Broad trigger phrases may start the workflow when the user intended a casual note rather than a file update.
\nMitigation: Narrow custom trigger phrases in Docs/CONFIG.md or ask the agent to confirm before writing workflow files.
\n\n\n## Reference(s):
\n- [ClawHub skill page](https://clawhub.ai/englandtong/daily-workflow)
\n- [Security Guide](SECURITY.md)
\n\n\n## Skill Output:
\n**Output Type(s):** [Text, Markdown, Configuration, Guidance]
\n**Output Format:** [Markdown files and concise text summaries]
\n**Output Parameters:** [1D]
\n**Other Properties Related to Output:** [Maintains project-local Docs/ workflow files; no global user state is written by default.]
\n\n## Skill Version(s):
\n3.0.0 (source: release evidence, SKILL.md, SECURITY.md, and _meta.json)
\n\n## Ethical Considerations:
\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\nFile v3.0.0:agents/openai.yaml\n\ndisplay_name: Daily Workflow / 项目记忆工作流\nshort_description: A bilingual project memory workflow that records project metadata, targets, current status, completed work, pending work, next actions, archive history, and AI handoff notes at project start, checkpoint, wrap-up, or handoff.\ndefault_prompt: Use this skill to initialize, migrate, or update project-local Docs/ workflow notes at start work, checkpoint, wrap-up, or handoff so the user and next AI can continue from clear project memory.\n\nArchive v2.0.0: 5 files, 6882 bytes\n\nFiles: _meta.json (133b), agents/openai.yaml (326b), SECURITY.md (1702b), skill-card.md (2372b), SKILL.md (9356b)\n\nFile v2.0.0:SKILL.md\n\n---\nname: daily-workflow\ndescription: \"Maintain project work-session status before, during, and after work. Use when the user says Chinese or English trigger phrases such as 开工啦/start work, 中段检查/checkpoint, 吃饭啦/lunch check, 收工啦/wrap up, or 交接/handoff, and needs Docs/ project notes updated for user visibility and AI-to-AI handoff.\"\n---\n\n# Daily Workflow Skill / 日常工作流 Skill\n\nVersion: 2.0.0\nPurpose: keep a project understandable across work sessions and across AI assistants.\nStorage: project-local `Docs/` only.\n\n## Core Principle / 核心原则\n\nWrite concise, factual project state that a user can scan and the next AI can continue from. Prefer concrete status, blockers, files, commands, and next actions over narrative.\n\nDo not record secrets, credentials, private customer data, or large sensitive source excerpts.\n\n## Context Compression and Reset / 上下文提炼与清理\n\nAt checkpoint, wrap-up, and handoff time, compress the current conversation context before updating workflow files.\n\nContext compression means extracting only the information needed to continue work:\n\n- Current state and user intent\n- Decisions already made\n- Completed work\n- Pending work\n- Blockers, risks, and assumptions\n- Files touched or important paths\n- Commands/tests run and their outcomes\n- Immediate next action\n\nWrite this compressed context into `Docs/STATUS.md` under the latest update entry. Do not paste the full conversation transcript.\n\nAfter saving the compressed context, treat the old working context as cleared for workflow purposes. Continue from the compressed context in `STATUS.md` and the action plan in `SCHEDULE.md`, not from long prior chat history.\n\nFor checkpoint only, provide the compressed context back to the current AI in the response so work can continue immediately from the reduced context.\n\n## Trigger Phrases / 触发语\n\nTreat these phrases as equivalent triggers:\n\n- Start work / 开工: `开工啦`, `开始工作`, `start work`, `begin work`\n- Mid-session checkpoint / 中段检查: `中段检查`, `吃饭啦`, `checkpoint`, `lunch check`, `midday check`\n- Wrap up / 收工: `收工啦`, `结束工作`, `wrap up`, `end work`\n- Handoff / 交接: `交接`, `handoff`, `next AI`, `下一任 AI`\n\nIf the user uses custom triggers documented in `Docs/CONFIG.md`, honor them.\n\n## Configuration / 配置\n\nOptional configuration is project-local:\n\n```text\nDocs/CONFIG.md\n```\n\nDefault config block:\n\n```json\n{\n \"version\": \"2.0.0\",\n \"startPhrase\": \"开工啦\",\n \"checkpointPhrase\": \"中段检查\",\n \"lunchPhrase\": \"吃饭啦\",\n \"endPhrase\": \"收工啦\",\n \"handoffPhrase\": \"交接\",\n \"language\": \"bilingual\",\n \"firstRun\": false\n}\n```\n\nCreate or update this config only inside `Docs/`. Do not write global user state.\n\n## Managed Files / 管理文件\n\nCreate `Docs/` if it is missing. Maintain these files:\n\n- `Docs/TARGET.md`: project purpose, success criteria, and scope. Default read-only after creation unless the user changes the goal.\n- `Docs/STATUS.md`: current state, progress, recent updates, and known risks. Update the current status section while preserving history.\n- `Docs/COMPLETED.md`: completed work log. Append only; never overwrite older entries.\n- `Docs/PENDING.md`: current pending work, blockers, and decisions needed. May be rewritten as the current queue, but preserve unresolved blockers.\n- `Docs/SCHEDULE.md`: immediate continuation plan and handoff block. Overwrite each checkpoint with the latest actionable next steps.\n\nOptional file:\n\n- `Docs/HANDOFF.md`: create only when the user asks for a standalone handoff file or when `SCHEDULE.md` becomes too long.\n- `Docs/CONFIG.md`: optional trigger and language preferences.\n\n## File Templates / 文件模板\n\n### TARGET.md\n\n```markdown\n# Project Target / 项目目标\n\n## Purpose / 目的\n[What the project is trying to accomplish.]\n\n## Success Criteria / 成功标准\n- [Criterion 1]\n- [Criterion 2]\n\n## Scope / 范围\nIn scope:\n- [Item]\n\nOut of scope:\n- [Item]\n\n## Last Reviewed / 最后确认\n[YYYY-MM-DD]\n```\n\n### STATUS.md\n\n```markdown\n# Project Status / 项目状态\n\n## Current State / 当前状态\n[Short factual summary.]\n\n## Progress / 进度\n[Percent or milestone-based progress when known.]\n\n## Active Context / 当前上下文\n- Files or areas being worked on:\n- Important constraints:\n- Known risks:\n\n## Compressed Context / 提炼上下文\n- User intent:\n- Decisions made:\n- Current state:\n- Completed work:\n- Pending work:\n- Blockers / risks:\n- Files touched:\n- Commands/tests run:\n- Immediate next action:\n\n## Update History / 更新历史\n### [YYYY-MM-DD HH:mm] [start/checkpoint/wrap-up/handoff]\n- [What changed]\n- [Evidence, command, or file reference when useful]\n```\n\n### COMPLETED.md\n\n```markdown\n# Completed Jobs / 已完成工作\n\n## [YYYY-MM-DD]\n- [Completed item with enough detail for another AI to understand what changed.]\n```\n\n### PENDING.md\n\n```markdown\n# Pending Jobs / 待办工作\n\n## Immediate / 立即处理\n- [ ] [Task, owner/context if known, acceptance condition]\n\n## Later / 稍后处理\n- [ ] [Task]\n\n## Blockers and Decisions / 阻塞与待决策\n- [Blocker or decision needed]\n```\n\n### SCHEDULE.md\n\n```markdown\n# Next Steps / 下一步\n\n## Immediate Next Action / 下一步立即行动\n1. [Action with file/path/context]\n2. [Action]\n\n## Handoff Summary / 交接摘要\n- Current state:\n- Completed since last checkpoint:\n- Pending work:\n- Blockers / risks:\n- Files touched:\n- Commands/tests run:\n- Immediate next action:\n\n## Notes for Next AI / 给下一任 AI 的说明\n[Any context the next AI must know before acting.]\n```\n\n## Start Work Flow / 开工流程\n\nWhen triggered by start work:\n\n1. Check `Docs/`; create missing managed files from templates.\n2. Check `Docs/CONFIG.md`; honor it if present, but do not require it.\n3. Read all managed files.\n4. Summarize the current project state to the user: target, current status, recent completed work, pending work, and immediate next action.\n5. If the target is missing or unclear, ask the user for the project goal before writing `TARGET.md`.\n6. Update `STATUS.md` with a start entry.\n7. Refresh `SCHEDULE.md` only if the existing next steps are missing, stale, or contradicted by the user.\n\n## Checkpoint Flow / 中段检查流程\n\nWhen triggered by checkpoint or lunch check:\n\n1. Compress the current conversation context into a concise continuation summary.\n2. Append completed items to `COMPLETED.md`.\n3. Update `STATUS.md` with current state, progress, risks, commands/tests run, and the compressed context.\n4. Rewrite `PENDING.md` as the current queue while preserving unresolved blockers.\n5. Rewrite `SCHEDULE.md` with a complete handoff block.\n6. Treat the previous long context as cleared for workflow purposes.\n7. Return the compressed context to the current AI in the response, followed by the immediate next action.\n\n## Wrap Up Flow / 收工流程\n\nWhen triggered by wrap up:\n\n1. Compress the current conversation context into a concise end-of-session summary.\n2. Append all completed work to `COMPLETED.md`.\n3. Update `STATUS.md` with final state, commands/tests run, risks, and the compressed context.\n4. Update `PENDING.md` with unfinished work, blockers, and decisions needed.\n5. Update `TARGET.md` only if the goal changed or the user explicitly asks.\n6. Rewrite `SCHEDULE.md` as the main handoff document.\n7. Treat the previous long context as cleared for workflow purposes.\n8. Give the user a short wrap-up summary and the immediate next action for a future session.\n\n## Handoff Flow / 交接流程\n\nWhen triggered by handoff:\n\n1. Read the managed files.\n2. Compress the current conversation context into a concise handoff summary.\n3. Update `STATUS.md` with the compressed context and latest observed state.\n4. Produce or refresh the `SCHEDULE.md` handoff block.\n5. Include only actionable context: current state, completed work since last checkpoint, pending tasks, blockers and risks, touched files, commands/tests run and outcomes, and immediate next action.\n6. Treat the previous long context as cleared for workflow purposes.\n7. Do not invent results that were not observed.\n\nCreate `Docs/HANDOFF.md` only if the user asks for a separate handoff artifact.\n\n## Write Rules / 写入规则\n\n- Use absolute paths when reporting file locations to the user.\n- Append to `COMPLETED.md`; do not overwrite it.\n- Preserve `STATUS.md` update history.\n- Preserve unresolved blockers when rewriting `PENDING.md`.\n- Overwrite `SCHEDULE.md` at each checkpoint or wrap-up so the next AI has a single current plan.\n- Never delete managed files automatically.\n- Do not edit files outside `Docs/` for this workflow unless the user's broader task requires it.\n\n## Safety Rules / 安全规则\n\nDo not write:\n\n- API keys, tokens, passwords, cookies, private keys, or `.env` values.\n- Full private customer records or confidential source text.\n- Browser session details or account credentials.\n- Large unrelated logs.\n\nIf important context contains sensitive details, summarize it safely and name the secure source location only when appropriate.\n\n## User Response Style / 用户反馈格式\n\nKeep responses short and useful:\n\n```text\nSaved workflow state.\n- Updated: STATUS.md, SCHEDULE.md\n- Appended: COMPLETED.md\n- Pending: 3 items, 1 blocker\n- Next: run the failing test and inspect the parser change\n```\n\nFile v2.0.0:_meta.json\n\n{\n \"ownerId\": \"kn7bjt2sd8f7cq83y0nf33xt19855gv6\",\n \"slug\": \"daily-workflow\",\n \"version\": \"2.0.0\",\n \"publishedAt\": 1778339993510\n}\n\nFile v2.0.0:SECURITY.md\n\n# Daily Workflow Skill - Security Guide\n\nVersion: 2.0.0\nLast updated: 2026-05-09\n\n## Security Statement\n\nThis skill maintains project-local status and handoff documents. It is intended to write only inside:\n\n```text\nDocs/\n```\n\nIt does not require network access, browser automation, shell access, or cloud upload.\n\n## Data Risks\n\nWorkflow notes can accidentally reveal:\n\n- Customer names or private business context\n- Internal project paths\n- Security findings before disclosure\n- Tokens, API keys, passwords, cookies, private keys, or `.env` values\n- Sensitive source snippets or logs\n\n## Required Protections\n\n- Do not record credentials or secret values.\n- Do not copy full private documents or long sensitive logs into handoff notes.\n- Summarize sensitive context instead of pasting it.\n- Do not delete managed workflow files automatically.\n- Do not write outside `Docs/` for workflow bookkeeping.\n- Preserve history in `COMPLETED.md` and `STATUS.md`.\n\n## File Handling\n\nAllowed write targets:\n\n- `Docs/TARGET.md`\n- `Docs/STATUS.md`\n- `Docs/COMPLETED.md`\n- `Docs/PENDING.md`\n- `Docs/SCHEDULE.md`\n- `Docs/HANDOFF.md` when explicitly useful or requested\n- `Docs/CONFIG.md` when custom trigger or language preferences are needed\n\nIf a path points outside the current project workspace, stop and ask the user before writing.\n\n## Release Checklist\n\n- `SKILL.md` and `_meta.json` show version `2.0.0`.\n- Text is valid UTF-8 and contains no mojibake.\n- Examples contain no real secrets or private paths.\n- The skill clearly states append vs overwrite rules.\n- Handoff format includes current state, completed work, pending work, blockers, files touched, commands/tests run, and immediate next action.\n\nFile v2.0.0:skill-card.md\n\n## Description:
\nMaintain project work-session status before, during, and after work through project-local workflow notes and handoff summaries.
\n\nThis skill is ready for commercial/non-commercial use.
\n\n## Publisher:
\n[englandtong](https://clawhub.ai/user/englandtong)
\n\n### License/Terms of Use:
\nMIT-0
\n\n\n## Use Case:
\nDevelopers and AI-assisted project teams use this skill to keep Docs/ workflow files current at start work, checkpoint, wrap-up, and handoff moments. It records concise state, pending work, blockers, commands, touched files, and next actions so the current user and next agent can continue with less context loss.
\n\n### Deployment Geography for Use:
\nGlobal
\n\n## Known Risks and Mitigations:
\nRisk: Workflow notes may accidentally include secrets, customer data, sensitive source excerpts, internal paths, or undisclosed security findings.
\nMitigation: Review Docs/ files before committing or sharing them, and summarize sensitive context instead of recording credentials, private records, long logs, or confidential source text.
\nRisk: The skill writes project-local workflow files and may overwrite current queues or next-step plans during checkpoint, wrap-up, or handoff.
\nMitigation: Keep the workflow scoped to Docs/, preserve history in STATUS.md and COMPLETED.md, and review PENDING.md and SCHEDULE.md after updates.
\n\n\n## Reference(s):
\n- [ClawHub release page](https://clawhub.ai/englandtong/daily-workflow)
\n- [Skill instructions](artifact/SKILL.md)
\n- [Security guide](artifact/SECURITY.md)
\n\n\n## Skill Output:
\n**Output Type(s):** [Markdown, Files, Guidance]
\n**Output Format:** [Markdown project notes and concise text summaries]
\n**Output Parameters:** [1D]
\n**Other Properties Related to Output:** [Creates or updates project-local Docs/ workflow files, including TARGET.md, STATUS.md, COMPLETED.md, PENDING.md, SCHEDULE.md, and optional HANDOFF.md or CONFIG.md.]
\n\n## Skill Version(s):
\n2.0.0 (source: server release metadata and artifact metadata)
\n\n## Ethical Considerations:
\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\nFile v2.0.0:agents/openai.yaml\n\ndisplay_name: Daily Workflow\nshort_description: Keep project status, checkpoints, and AI handoff notes current during a work session.\ndefault_prompt: Use this skill to initialize or update Docs/ project workflow notes at start work, checkpoint, wrap-up, or handoff so the user and next AI can continue with clear context.\n\nArchive v1.1.0: 2 files, 3688 bytes\n\nFiles: SKILL.md (9051b), _meta.json (133b)\n\nFile v1.1.0:SKILL.md\n\n---\r\nname: daily-workflow\r\ndescription: \"Daily work start, mid-day check-in, and end workflow manager. Triggered by customizable phrases (default: '开工啦', '吃饭啦', '收工啦'). Manages project documentation (Docs/ folder): PROJECT_TARGET.md, PROJECT_STATUS.md, COMPLETED_JOBS.md, PENDING_JOBS.md, NEXT_STEPS.md. Supports AI-to-AI project handover with clear context.\"\r\nagent_created: true\r\n---\r\n\r\n# Daily Workflow Skill (中文版)\r\n\r\n管理工作会话，自动化项目文档追踪，支持 AI 间项目交接。\r\n\r\n## 首次使用配置\r\n\r\n当技能首次被触发时，检查配置文件 `.workbuddy/daily-workflow-config.json` 是否存在。\r\n\r\n如果配置文件不存在，执行以下初始化流程：\r\n\r\n1. **询问用户自定义暗语**\r\n - 询问用户希望使用什么暗语来触发三个工作流节点\r\n - 提供默认建议：\r\n - 开工暗语：开工啦\r\n - 吃饭暗语：吃饭啦\r\n - 收工暗语：收工啦\r\n - 允许用户自定义每个暗语\r\n - 支持中英文混合\r\n\r\n2. **创建配置文件**\r\n - 在项目根目录的 `.workbuddy/` 文件夹中创建 `daily-workflow-config.json`\r\n - 文件格式：\r\n ```json\r\n {\r\n \"startPhrase\": \"开工啦\",\r\n \"lunchPhrase\": \"吃饭啦\",\r\n \"endPhrase\": \"收工啦\",\r\n \"language\": \"zh-CN\",\r\n \"firstRun\": false\r\n }\r\n ```\r\n\r\n3. **确认配置**\r\n - 向用户展示配置结果\r\n - 说明可以随时修改配置文件或重新运行配置\r\n\r\n如果配置文件已存在，直接读取配置并使用用户设定的暗语。\r\n\r\n## 技能触发时机\r\n\r\n根据配置文件中的暗语触发技能，默认触发词：\r\n- **\"开工啦\"** - 开始工作日\r\n- **\"吃饭啦\"** - 工作中途检查点（午休前）\r\n- **\"收工啦\"** - 结束工作日\r\n\r\n这些短语标志著有意图的工作会话边界，应该捕获和恢复项目状态。\r\n\r\n## 核心工作流\r\n\r\n### Phase 1: 开工 (\"开工啦\")\r\n\r\n当用户表示开始工作时，执行以下流程：\r\n\r\n1. **确保 Docs 目录存在**\r\n - 检查工作区根目录是否存在 `Docs/` 目录\r\n - 如果不存在，创建 `Docs/`\r\n\r\n2. **检查并创建所需文件**\r\n - 验证 `Docs/` 中是否存在以下五个文件：\r\n - `PROJECT_TARGET.md` - 项目目标和目的\r\n - `PROJECT_STATUS.md` - 当前项目状态\r\n - `COMPLETED_JOBS.md` - 已完成的工作项目\r\n - `PENDING_JOBS.md` - 待处理/未完成的工作项目\r\n - `NEXT_STEPS.md` - 计划的下一步\r\n - 创建任何缺失的文件，使用适当的模板标头\r\n\r\n3. **读取所有文档**\r\n - 读取 `Docs/` 中的所有五个文件\r\n - 提取关键信息：\r\n - 当前项目目标\r\n - 最新项目状态\r\n - 最近完成的工作\r\n - 待处理任务和支持问题\r\n - 计划的下一步\r\n\r\n4. **展示工作会话简报**\r\n - 总结当前项目状态\r\n - 强调自上次会话以来完成的工作\r\n - 识别待处理任务\r\n - 建议当前会话的优先工作\r\n - 询问用户确认或调整计划\r\n\r\n### Phase 1.5: 吃饭啦（工作中途检查点）\r\n\r\n当用户说\"吃饭啦\"（或自定义暗语）时，执行以下流程：\r\n\r\n**目的：** 在午休或长时间中断前，更新项目状态，确保回来后能快速恢复工作。\r\n\r\n1. **更新项目状态**\r\n - 更新 `Docs/PROJECT_STATUS.md`，记录：\r\n - 当前完成百分比\r\n - 正在进行的任务\r\n - 当前焦点领域\r\n - 任何项目方向的变化\r\n\r\n2. **更新已完成工作**\r\n - 将上午完成的工作项目追加到 `Docs/COMPLETED_JOBS.md`\r\n - 使用格式：`## [YYYY-MM-DD 上午]\\n- [已完成项目1]\\n- [已完成项目2]`\r\n - 包含足够细节，让另一个 AI 能理解做了什么\r\n\r\n3. **更新待处理工作**\r\n - 更新 `Docs/PENDING_JOBS.md`，记录：\r\n - 已开始但未完成的任务\r\n - 遇到的阻碍\r\n - 推迟到下午的任务\r\n - 从待处理列表中移除已完成项目\r\n\r\n4. **更新下一步计划**\r\n - 写入清晰的、可操作的下一步到 `Docs/NEXT_STEPS.md`\r\n - 包含：\r\n - 下午回来后要处理的具体任务\r\n - 优先顺序\r\n - 任何先决条件或依赖关系\r\n - 立即恢复工作所需的上下文\r\n\r\n5. **展示中途检查摘要**\r\n - 总结上午完成的工作\r\n - 列出更新的文档文件\r\n - 确认下午的下一步已清楚记录\r\n - 确保回来工作时能快速恢复状态\r\n\r\n### Phase 2: 收工 (\"收工啦\")\r\n\r\n当用户表示结束工作时，执行以下流程：\r\n\r\n1. **更新已完成工作**\r\n - 将今天完成的工作项目追加到 `Docs/COMPLETED_JOBS.md`\r\n - 使用格式：`## [YYYY-MM-DD]\\n- [已完成项目1]\\n- [已完成项目2]`\r\n - 包含足够细节，让另一个 AI 能理解做了什么\r\n\r\n2. **更新待处理工作**\r\n - 更新 `Docs/PENDING_JOBS.md`，包含：\r\n - 已开始但未完成的任务\r\n - 遇到的阻碍\r\n - 推迟到下次会话的任务\r\n - 从待处理列表中移除已完成项目\r\n\r\n3. **更新项目状态**\r\n - 更新 `Docs/PROJECT_STATUS.md`，包含：\r\n - 当前完成百分比\r\n - 达到的关键里程碑\r\n - 当前焦点领域\r\n - 任何项目方向的变化\r\n\r\n4. **更新项目目标（如需要）**\r\n - 仅在以下情况修改 `Docs/PROJECT_TARGET.md`：\r\n - 项目目标已变更\r\n - 发现新需求\r\n - 需要调整范围\r\n - 否则，保持不变\r\n\r\n5. **更新下一步**\r\n - 写入清晰的、可操作的下一步到 `Docs/NEXT_STEPS.md`\r\n - 包含：\r\n - 下次会话要处理的具体任务\r\n - 优先顺序\r\n - 任何先决条件或依赖关系\r\n - 立即恢复工作所需的上下文\r\n\r\n6. **展示会话摘要**\r\n - 总结完成的工作\r\n - 列出更新的文档文件\r\n - 确认下一步已清楚记录\r\n - 确保为下一个 AI 准备好交接状态\r\n\r\n## 文件模板\r\n\r\n创建缺失文件时，使用以下模板：\r\n\r\n### PROJECT_TARGET.md\r\n```markdown\r\n# 项目目标\r\n\r\n## 项目概述\r\n[描述项目的主要目标]\r\n\r\n## 关键目标\r\n- [目标1]\r\n- [目标2]\r\n\r\n## 成功标准\r\n- [标准1]\r\n- [标准2]\r\n\r\n## 最后更新\r\n[YYYY-MM-DD]\r\n```\r\n\r\n### PROJECT_STATUS.md\r\n```markdown\r\n# 项目状态\r\n\r\n## 当前状态\r\n[当前状态的简要描述]\r\n\r\n## 完成度\r\n[XX]% 完成\r\n\r\n## 当前焦点\r\n[现在正在做什么]\r\n\r\n## 最近里程碑\r\n- [YYYY-MM-DD] [里程碑描述]\r\n\r\n## 最后更新\r\n[YYYY-MM-DD]\r\n```\r\n\r\n### COMPLETED_JOBS.md\r\n```markdown\r\n# 已完成工作\r\n\r\n## [YYYY-MM-DD]\r\n- [已完成任务1]\r\n- [已完成任务2]\r\n\r\n## [YYYY-MM-DD 上午]\r\n- [上午完成的任务1]\r\n\r\n## [YYYY-MM-DD]\r\n- [已完成任务1]\r\n```\r\n\r\n### PENDING_JOBS.md\r\n```markdown\r\n# 待处理工作\r\n\r\n## 高优先级\r\n- [ ] [任务1]\r\n- [ ] [任务2]\r\n\r\n## 中优先级\r\n- [ ] [任务3]\r\n\r\n## 低优先级\r\n- [ ] [任务4]\r\n\r\n## 阻碍\r\n- [阻碍描述（如有）]\r\n```\r\n\r\n### NEXT_STEPS.md\r\n```markdown\r\n# 下一步\r\n\r\n## 立即行动（下次会话）\r\n1. [行动1 - 附上下文]\r\n2. [行动2 - 附上下文]\r\n\r\n## 即将到来的任务\r\n- [任务1]\r\n- [任务2]\r\n\r\n## 给下一个 AI 的笔记\r\n[重要上下文、做出的决定、需要记住的事情]\r\n```\r\n\r\n## AI 到 AI 交接原则\r\n\r\n写入所有文档时，假设下一个读者将是一个不同的 AI 实例，它需要：\r\n- 理解正在构建什么以及为什么\r\n- 无需询问基本问题即可恢复工作\r\n- 继续相同的编码风格和约定\r\n- 尊重已做出的决定\r\n\r\n**关键实践：**\r\n- 详细写入，而非简写\r\n- 解释\"为什么\"而不仅仅是\"是什么\"\r\n- 在相关时包含代码片段或文件引用\r\n- 记录任何变通方法、技巧或技术债务\r\n- 记录用户偏好和决定\r\n\r\n## 工作流图表\r\n\r\n```\r\n用户说\"开工啦\"\r\n ↓\r\n检查 Docs/ 是否存在 → 如缺失则创建\r\n ↓\r\n检查 5 个文件是否存在 → 创建缺失的模板\r\n ↓\r\n读取所有 5 个文件\r\n ↓\r\n展示工作会话简报\r\n ↓\r\n用户与 AI 协助工作\r\n ↓\r\n用户说\"吃饭啦\"（午休检查点）\r\n ↓\r\n更新 PROJECT_STATUS.md\r\n ↓\r\n更新 COMPLETED_JOBS.md\r\n ↓\r\n更新 PENDING_JOBS.md\r\n ↓\r\n更新 NEXT_STEPS.md\r\n ↓\r\n展示中途检查摘要\r\n ↓\r\n用户午休后返回继续工作\r\n ↓\r\n用户说\"收工啦\"\r\n ↓\r\n更新 COMPLETED_JOBS.md\r\n ↓\r\n更新 PENDING_JOBS.md\r\n ↓\r\n更新 PROJECT_STATUS.md\r\n ↓\r\n更新 PROJECT_TARGET.md（如需要）\r\n ↓\r\n更新 NEXT_STEPS.md\r\n ↓\r\n展示会话摘要\r\n ↓\r\n项目状态已保存，供下一个 AI 使用\r\n```\r\n\r\n## 重要注意事项\r\n\r\n- 引用 `Docs/` 目录时始终使用绝对路径\r\n- 追加到 `COMPLETED_JOBS.md`（绝不覆盖先前的条目）\r\n- 每次会话覆盖 `NEXT_STEPS.md`（它是为紧接的下一个会话准备的）\r\n- 在 `PROJECT_STATUS.md` 和 `COMPLETED_JOBS.md` 中保留历史信息\r\n- 如有疑问，写入更多上下文，而非更少\r\n- 首次使用时会询问用户自定义暗语，后续使用已保存的配置\r\n- 配置文件位置：`.workbuddy/daily-workflow-config.json`\n\nFile v1.1.0:_meta.json\n\n{\n \"ownerId\": \"kn7bjt2sd8f7cq83y0nf33xt19855gv6\",\n \"slug\": \"daily-workflow\",\n \"version\": \"1.1.0\",\n \"publishedAt\": 1777996947789\n}\n\nArchive v1.0.0: 2 files, 2830 bytes\n\nFiles: SKILL.md (6323b), _meta.json (133b)\n\nFile v1.0.0:SKILL.md\n\n---\r\nname: daily-workflow\r\ndescription: \"Daily work start and end workflow manager. Triggered by phrases like 'kai gong la' (starting work) or 'shou gong la' (ending work). Manages project documentation (Docs/ folder): PROJECT_TARGET.md, PROJECT_STATUS.md, COMPLETED_JOBS.md, PENDING_JOBS.md, NEXT_STEPS.md. Ensures AI-to-AI project handover with clear context.\"\r\nagent_created: true\r\n---\r\n\r\n# Daily Workflow Skill\r\n\r\nManage daily work sessions with automated project documentation tracking and AI-to-AI handover support.\r\n\r\n## When to Use This Skill\r\n\r\nTrigger this skill when the user mentions:\r\n- **\"开工啦\"** - Starting the work day\r\n- **\"收工啦\"** - Ending the work day\r\n\r\nThese phrases indicate intentional work session boundaries where project state should be captured and restored.\r\n\r\n## Core Workflow\r\n\r\n### Phase 1: Start Work (\"开工啦\")\r\n\r\nExecute the following sequence when the user indicates work is starting:\r\n\r\n1. **Ensure Docs Directory exists**\r\n - Check if `Docs/` directory exists in workspace root\r\n - Create `Docs/` if it does not exist\r\n\r\n2. **Check and Create Required Files**\r\n - Verify these five files exist in `Docs/`:\r\n - `PROJECT_TARGET.md` - Project objectives and goals\r\n - `PROJECT_STATUS.md` - Current project status\r\n - `COMPLETED_JOBS.md` - Completed work items\r\n - `PENDING_JOBS.md` - Pending/incomplete work items\r\n - `NEXT_STEPS.md` - Planned next steps\r\n - Create any missing files with appropriate template headers\r\n\r\n3. **Read All Documentation**\r\n - Read all five files from `Docs/`\r\n - Extract key information:\r\n - Current project objectives\r\n - Latest project status\r\n - Recently completed work\r\n - Pending tasks and blockers\r\n - Planned next steps\r\n\r\n4. **Present Work Session Briefing**\r\n - Summarize current project state\r\n - Highlight completed work since last session\r\n - Identify pending tasks\r\n - Suggest priority work for current session\r\n - Ask user for confirmation or adjustments to the plan\r\n\r\n### Phase 2: End Work (\"收工啦\")\r\n\r\nExecute the following sequence when the user indicates work is ending:\r\n\r\n1. **Update Completed Jobs**\r\n - Append today's completed work items to `Docs/COMPLETED_JOBS.md`\r\n - Use format: `## [YYYY-MM-DD]\\n- [completed item 1]\\n- [completed item 2]`\r\n - Include sufficient detail for another AI to understand what was done\r\n\r\n2. **Update Pending Jobs**\r\n - Update `Docs/PENDING_JOBS.md` with:\r\n - Tasks started but not completed\r\n - Blockers encountered\r\n - Tasks deferred to next session\r\n - Remove completed items from pending list\r\n\r\n3. **Update Project Status**\r\n - Update `Docs/PROJECT_STATUS.md` with:\r\n - Current completion percentage\r\n - Key milestones reached\r\n - Current focus area\r\n - Any changes in project direction\r\n\r\n4. **Update Project Target (if needed)**\r\n - Modify `Docs/PROJECT_TARGET.md` only if:\r\n - Project objectives have changed\r\n - New requirements discovered\r\n - Scope adjustments needed\r\n - Otherwise, leave unchanged\r\n\r\n5. **Update Next Steps**\r\n - Write clear, actionable next steps to `Docs/NEXT_STEPS.md`\r\n - Include:\r\n - Specific tasks to tackle next session\r\n - Priority order\r\n - Any prerequisites or dependencies\r\n - Context needed to resume work immediately\r\n\r\n6. **Present Session Summary**\r\n - Summarize what was completed\r\n - List updated documentation files\r\n - Confirm next steps are clearly documented\r\n - Ensure handover-ready state for next AI\r\n\r\n## File Templates\r\n\r\nWhen creating missing files, use these templates:\r\n\r\n### PROJECT_TARGET.md\r\n```markdown\r\n# Project Target\r\n\r\n## Project Overview\r\n[Describe the project's main objective]\r\n\r\n## Key Goals\r\n- [Goal 1]\r\n- [Goal 2]\r\n\r\n## Success Criteria\r\n- [Criterion 1]\r\n- [Criterion 2]\r\n\r\n## Last Updated\r\n[YYYY-MM-DD]\r\n```\r\n\r\n### PROJECT_STATUS.md\r\n```markdown\r\n# Project Status\r\n\r\n## Current Status\r\n[Brief description of current state]\r\n\r\n## Completion\r\n[XX]% complete\r\n\r\n## Current Focus\r\n[What is being worked on now]\r\n\r\n## Recent Milestones\r\n- [YYYY-MM-DD] [Milestone description]\r\n\r\n## Last Updated\r\n[YYYY-MM-DD]\r\n```\r\n\r\n### COMPLETED_JOBS.md\r\n```markdown\r\n# Completed Jobs\r\n\r\n## [YYYY-MM-DD]\r\n- [Completed task 1]\r\n- [Completed task 2]\r\n\r\n## [YYYY-MM-DD]\r\n- [Completed task 1]\r\n```\r\n\r\n### PENDING_JOBS.md\r\n```markdown\r\n# Pending Jobs\r\n\r\n## High Priority\r\n- [ ] [Task 1]\r\n- [ ] [Task 2]\r\n\r\n## Medium Priority\r\n- [ ] [Task 3]\r\n\r\n## Low Priority\r\n- [ ] [Task 4]\r\n\r\n## Blockers\r\n- [Blocker description if any]\r\n```\r\n\r\n### NEXT_STEPS.md\r\n```markdown\r\n# Next Steps\r\n\r\n## Immediate Actions (Next Session)\r\n1. [Action 1 - with context]\r\n2. [Action 2 - with context]\r\n\r\n## Upcoming Tasks\r\n- [Task 1]\r\n- [Task 2]\r\n\r\n## Notes for Next AI\r\n[Important context, decisions made, things to remember]\r\n```\r\n\r\n## AI-to-AI Handover Principles\r\n\r\nWrite all documentation assuming the next reader will be a different AI instance that needs to:\r\n- Understand what was being built and why\r\n- Resume work without asking basic questions\r\n- Continue the same coding style and conventions\r\n- Respect decisions already made\r\n\r\n**Key practices:**\r\n- Write in detail, not shorthand\r\n- Explain the \"why\" not just the \"what\"\r\n- Include code snippets or file references when relevant\r\n- Note any workarounds, hacks, or technical debt\r\n- Record user preferences and decisions\r\n\r\n## Workflow Diagram\r\n\r\n```\r\nUser says \"开工啦\"\r\n ↓\r\nCheck Docs/ exists → Create if missing\r\n ↓\r\nCheck 5 files exist → Create missing with templates\r\n ↓\r\nRead all 5 files\r\n ↓\r\nPresent work session briefing\r\n ↓\r\nUser works with AI assistance\r\n ↓\r\nUser says \"收工啦\"\r\n ↓\r\nUpdate COMPLETED_JOBS.md\r\n ↓\r\nUpdate PENDING_JOBS.md\r\n ↓\r\nUpdate PROJECT_STATUS.md\r\n ↓\r\nUpdate PROJECT_TARGET.md (if needed)\r\n ↓\r\nUpdate NEXT_STEPS.md\r\n ↓\r\nPresent session summary\r\n ↓\r\nProject state saved for next AI\r\n```\r\n\r\n## Important Notes\r\n\r\n- Always use absolute paths when referencing `Docs/` directory\r\n- Append to `COMPLETED_JOBS.md` (never overwrite previous entries)\r\n- Overwrite `NEXT_STEPS.md` each session (it's for the immediate next session)\r\n- Preserve historical information in `PROJECT_STATUS.md` and `COMPLETED_JOBS.md`\r\n- When in doubt, write more context, not less\n\nFile v1.0.0:_meta.json\n\n{\n \"ownerId\": \"kn7bjt2sd8f7cq83y0nf33xt19855gv6\",\n \"slug\": \"daily-workflow\",\n \"version\": \"1.0.0\",\n \"publishedAt\": 1777868597746\n}","readmeExcerpt":"Skill: Daily Workflow Owner: englandtong Summary: Preserve evidence-backed project memory Tags: ai-handoff:4.1.0, ai-handover:4.1.0, ai-workflow:3.0.0, bilingual:4.1.0, checkpoint:4.1.0, chinese:4.1.0, context-compression:4.1.0, context-management:4.1.0, daily-routine:4.1.0, daily-standup:4.1.0, daily-workflow:4.1.0, docs:2.0.0, documentation:4.1.0, en:4.1.0, english:4.1.0, handoff:4.1.0, latest:4.1.0, productivity:4","codeSnippets":[],"executableExamples":[{"language":"text","snippet":"Docs/STATUS.md\nDocs/NEXT_ACTIONS.md"},{"language":"text","snippet":"Docs/PROJECT.md project identity and metadata\nDocs/TARGET.md Owner-confirmed target, scope, Non-Goals, success criteria\nDocs/COMPLETED.md append-only historical completion log\nDocs/PENDING.md current queue, blockers, and decisions\nDocs/HANDOFF.md standalone handoff only when requested or too large to embed\nDocs/CONFIG.md explicit local workflow preferences\nDocs/archive/YYYY-MM.md"},{"language":"text","snippet":"PROJECT_TARGET.md -> TARGET.md\nPROJECT_STATUS.md -> STATUS.md\nCOMPLETED_JOBS.md -> COMPLETED.md\nPENDING_JOBS.md -> PENDING.md\nNEXT_STEPS.md -> NEXT_ACTIONS.md\nSCHEDULE.md -> NEXT_ACTIONS.md compatibility alias"},{"language":"markdown","snippet":"# Project Status\n\n## Current State\n- Status:\n- Current goal:\n- Active scope:\n- Latest verified behavior:\n- Blockers:\n- Residual risks:\n\n## Latest Verification\n- Command or scenario:\n- Final result:\n- Evidence boundary:\n- Evidence path:\n- Not executed / deferred:\n\n## Compressed Context\n- User intent:\n- Decisions:\n- Completed since last checkpoint:\n- Pending:\n- Files touched:\n- Dirty worktree notes:\n- Exact next action:\n\n## Update History\n### YYYY-MM-DD HH:mm [start|checkpoint|wrap-up|handoff|migration]\n- Factual change and evidence."},{"language":"markdown","snippet":"# Next Actions\n\n## Immediate Next Action\n1. [Action with context and acceptance condition]\n\n## Then\n1. [Ordered action]\n\n## Blockers And Owner Decisions\n- [Decision or blocker]\n\n## Handoff Snapshot\n- Current state:\n- Completed:\n- Pending:\n- Risks:\n- Files/artifacts:\n- Commands and final results:\n- Exact next action:"},{"language":"text","snippet":"Project memory updated.\n- Current state: Needs Fix\n- Verified: 2 checks passed; browser flow not executed\n- Updated: STATUS.md, NEXT_ACTIONS.md\n- Preserved: active QA and loop evidence\n- Blocker: Owner decision on target scope\n- Next: reproduce the failing user-visible flow with the current data boundary"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: daily-workflow\ndescription: Say 开工啦 or 收工啦 and get a resumable project note written for you, so tomorrow you or another AI can pick up without re-reading everything. Use when a user asks to resume a project, save progress, record a checkpoint, end a work session, prepare a handoff, switch to a new AI session, or reconcile stale working notes. Typical triggers include 开工啦, 中段检查, 吃饭啦, 收工啦, 交接, 记一下进度, 存个档, 明天接着做, 换个 AI 继续, 上下文快满了, start work, checkpoint, wrap up, handoff, save progress, resume tomorrow, and switch to a new agent session. Reads existing project authority first, preserves dirty worktrees and governance evidence, avoids competing state files, archives rather than deletes history, and records commands, final outcomes, blockers, risks, and exact next actions without claiming unverified completion or QA acceptance.\n---\n\n# Daily Workflow / 项目记忆工作流\n\nVersion: 4.1.0\n\nUse this skill to make work resumable. Record only the compact factual state needed by the user or next agent; do not turn project memory into a second project-management system.\n\nRespond in the user's language. Keep machine-readable status values in English.\n\n## Scope And Ownership\n\nThis skill owns session continuity only: orientation, checkpoint summaries, wrap-up records, and handoffs.\n\n- `cms-project-governance` owns formal targets, Programs, Work Orders, Controller/QA decisions, and acceptance state.\n- `agent-loop-engineering` owns active coding-loop evidence, stop gates, evaluation, and loop records.\n- `project-lifecycle-navigator` owns lifecycle analysis and rebaseline proposals.\n- `web-search-rules` owns research intake and source rules.\n- `ai-workflow-os` may route here but must not create a parallel memory state.\n\nWhen another system already owns a field, read and summarize it; do not overwrite it.\n\n## Trigger And Authorization\n\nUse this skill when the user explicitly asks to:\n\n- start or resume work;\n- save progress or create a checkpoint;\n- wrap up or end work;\n- prepare a handoff;\n- reconcile or migrate workflow notes.\n\nCommon phrases include `开工啦`, `中段检查`, `吃饭啦`, `收工啦`, `交接`, `start work`, `checkpoint`, `wrap up`, and `handoff`.\n\nA casual mention of one phrase is not enough when intent is ambiguous. Inspect read-only first. Persist files only when the request clearly asks to initialize or update project memory, or when an established project workflow already defines the trigger as a write command.\n\n## Read-Only Orientation First\n\nBefore writing:\n\n1. resolve the actual workspace and repository root, including nested repositories;\n2. inspect Git branch, commit, and dirty state without modifying it;\n3. locate governing files and determine current authority precedence;\n4. locate product scope, entrypoints, build/test commands, and runtime surfaces relevant to the session;\n5. read existing memory files and compatible aliases;\n6. identify contradictions, missing evidence, and ownership conflicts.\n\nDo not use `git reset`, `git clean`, broad deletion, or ambiguous"},{"path":"_meta.json","content":"{\n \"ownerId\": \"kn7bjt2sd8f7cq83y0nf33xt19855gv6\",\n \"slug\": \"daily-workflow\",\n \"version\": \"4.1.0\",\n \"publishedAt\": 1788800774694\n}"},{"path":"QUICKSTART.md","content":"# Quickstart / 快速开始\n\n## Install / 安装\n\n```bash\nopenclaw skills install daily-workflow\n```\n\nor / 或\n\n```bash\nclawhub install daily-workflow\n```\n\n## Try it in 30 seconds / 30 秒验证\n\nSay this to your agent at the end of a work session / 收工时对 agent 说：\n\n```text\n收工啦\n```\n\nYou should get / 你应该得到：\n\n- `Docs/STATUS.md` — current state, latest verification, next action;\n- `Docs/NEXT_ACTIONS.md` — the immediate next step, then what follows,\n blockers and Owner decisions;\n- a one-line summary of what was written.\n\nOpen those two files. That is the entire output — no dashboard, no database.\n\n## Four phrases are enough / 四句话就够\n\n| Say | What happens |\n| --- | --- |\n| `开工啦` | read existing state, orient for the session |\n| `中段检查` | mid-session checkpoint |\n| `收工啦` | wrap up, record exact next actions |\n| `交接` | package a handoff for another person or another AI |\n\nThe point is resumability: tomorrow you, or a fresh AI session, can pick up\nwithout re-reading the whole project.\n\n## Minimum useful path / 最小可用路径\n\nJust use `开工啦` and `收工啦`.\n\nYou do **not** need `Docs/PROJECT.md`, `TARGET.md`, `COMPLETED.md` or\n`PENDING.md`. Those belong to the governance profile and only appear if your\nproject already runs one. The default lightweight profile writes two files.\n\n## Privacy and safety / 隐私与安全\n\n- Runs locally. Writes plain Markdown into your project's own `Docs/`.\n- Stores no API keys, tokens, passwords or credentials.\n- Never deletes history — superseded entries are archived, not removed.\n- Read-only orientation first: nothing is written until intent is clear.\n- Will not claim work is complete or QA-accepted without evidence.\n\n## Next / 下一步\n\nFull record structure and memory profiles: `SKILL.md`."},{"path":"SECURITY.md","content":"# Daily Workflow / 项目记忆工作流 - Security Guide\n\nVersion: 4.1.0\n\n## Security Statement / 安全说明\n\nThis skill may write project-local workflow notes when the user explicitly requests persistence or an established trigger authorizes it. It is designed for resumability and handoff, not for storing secrets, replacing project governance, or recording confidential source bodies.\n\n本 Skill 只应在项目本地 `Docs/` 中写入工作流笔记，用于恢复上下文和交接，不用于保存密钥、账号凭证或完整敏感资料。\n\n## Allowed Storage / 允许写入范围\n\nDefault allowed write scope when no project-owned workflow defines a narrower location:\n\n```text\nDocs/\n```\n\nDo not write global user state. Do not write outside `Docs/` unless the user explicitly asks as part of the broader task.\n\nBefore writing, resolve the actual workspace and repository, preserve dirty changes, identify existing governance ownership, and avoid creating competing state files.\n\n## Sensitive Data Rules / 敏感信息规则\n\nNever record:\n\n- API keys, tokens, passwords, private keys, cookies, OAuth refresh tokens, browser sessions.\n- `.env` values or credential files.\n- Full private customer records.\n- Long confidential source excerpts.\n- Large unrelated logs.\n\nIf sensitive context is relevant, record a safe summary and the non-secret location where the user can find the source.\n\n## Legacy Migration Safety / 旧版迁移安全\n\nLegacy files may be read for migration, but must not be deleted automatically. If both old and new files exist, ask before merging. Do not migrate secret-like fields.\n\n## Archive Safety / 归档安全\n\nArchiving older status history is allowed only by moving content into `Docs/archive/`. Never delete history automatically.\n\n## Release Checklist / 发布检查\n\n- `SKILL.md` and `SECURITY.md` show version `4.1.0`; generated registry snapshots are excluded from the publish bundle.\n- The package uses one bilingual skill, not separate English and Chinese packages.\n- Legacy file migration is documented.\n- `NEXT_ACTIONS.md` is the primary continuation file, with `SCHEDULE.md` treated as compatibility alias.\n- Existing project/governance files remain authoritative; the lightweight profile starts with only `STATUS.md` and `NEXT_ACTIONS.md`.\n- Completion, verification, and independent acceptance remain separate.\n- No example contains real credentials or private data."},{"path":"skill-card.md","content":"## Description:\n\nDaily Workflow helps agents create evidence-backed project checkpoints, wrap-ups, and handoffs so work can resume without re-reading the whole project.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[englandtong](https://clawhub.ai/user/englandtong)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and agent users use this skill to preserve concise project memory at start, checkpoint, wrap-up, and handoff moments. It writes project-local Markdown notes that capture current state, verification evidence, blockers, risks, and exact next actions without replacing existing governance records.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Checkpoint or handoff notes can contain sensitive project context if shared outside the project.\n\nMitigation: Review generated Docs/ notes before sharing and avoid recording secrets, credentials, full confidential data, or large private logs.\n\nRisk: Project memory files can conflict with existing governance or acceptance records.\n\nMitigation: Read existing project authority first, preserve dirty worktrees and governance evidence, and summarize externally owned fields instead of overwriting them.\n\n## Reference(s):\n\n- [ClawHub Skill Page](https://clawhub.ai/englandtong/skills/daily-workflow)\n- [Quickstart](artifact/QUICKSTART.md)\n- [Security Guide](artifact/SECURITY.md)\n\n## Skill Output:\n\n**Output Type(s):** [Markdown, Files, Guidance]\n\n**Output Format:** [Markdown files and concise user-facing summaries]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Writes project-local workflow notes, normally under Docs/, when persistence is clearly authorized.]\n\n## Skill Version(s):\n\n4.1.0 (source: server release metadata and artifact SKILL.md)\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":null,"editorialQuality":{"score":100,"threshold":65,"status":"thin","wordCount":1782,"uniquenessScore":44,"reasons":["uniqueness-below-45"]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-10T21:03:35.764Z","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":"2026-10-10T21:03:35.764Z","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":"2026-10-10T23:47:03.515Z","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"}]}}}