[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6R7FGP3R612506A5GKMBF5C":3,"skill-same:01M6R7FGP3R612506A5GKMBF5C":86},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":57,"commitSha":58,"skillDirectory":59,"forksCount":60,"repositoryHtmlUrl":61,"repositoryHomepageUrl":6,"ownerHtmlUrl":62,"ownerBlogUrl":63,"ownerDescription":64,"updatedAt":65,"install":66,"resources":79},"01M6R7FGP3R612506A5GKMBF5C","harness-schedule-run",null,"Use when running Plans.md tasks on a scheduled cadence (fresh context per wake-up, sprint-contract flow, plateau detection, flock guard). Do NOT load for: single-task work (harness-work), planning, review, release.","tim-hub\u002Fpowerball-harness","tim-hub","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F2884320?v=4","01M6R7B70M93Y25SH3DV1B4Q26","Tim","01M6R7B74YQWC556YVC0FRVS3T",35,"MIT","https:\u002F\u002Fgithub.com\u002Ftim-hub\u002Fpowerball-harness\u002Fblob\u002Fmaster\u002Fharness\u002Ftemplates\u002Fcodex-skills\u002Fharness-schedule-run\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"harness schedule run","run","harness","schedule","mem","when","sprint-contract","wake","scheduled","cadence","sprint","fresh","worker","MCP","up","work","Bash","ci","tasks","review","2026-11-10 10:12:30",false,"---\nname: harness-schedule-run\ndescription: \"Use when running Plans.md tasks on a scheduled cadence (fresh context per wake-up, sprint-contract flow, plateau detection, flock guard). Do NOT load for: single-task work (harness-work), planning, review, release.\"\nallowed-tools: [\"Read\", \"Edit\", \"Bash\", \"Task\", \"ScheduleWakeup\", \"mcp__harness__harness_mem_resume_pack\", \"mcp__harness__harness_mem_record_checkpoint\"]\nargument-hint: \"[all|N-M] [--max-cycles N] [--pacing worker|ci|plateau|night]\"\n---\n\n# Harness Schedule Run\n\nMeta-skill that combines `\u002Floop` (CC dynamic mode) with `ScheduleWakeup` to re-enter long-running tasks with a **fresh context on every wake-up**.\n\nEach wake-up calls `harness-work --breezing` via the Agent tool, forming a re-entrant loop of 1 cycle = 1 task completion.\n\n## Quick Reference\n\n| Input | Behavior |\n|-------|----------|\n| `\u002Fharness-schedule-run all` | Loop all incomplete tasks (default: max 8 cycles) |\n| `\u002Fharness-schedule-run all --max-cycles 3` | Stop after 3 cycles |\n| `\u002Fharness-schedule-run 41.1-41.3 --pacing ci` | Execute task range with CI pacing |\n| `\u002Fharness-schedule-run all --pacing night` | Overnight batch (3600s interval) |\n\n## Options\n\n| Option | Description | Default |\n|--------|-------------|---------|\n| `all` | Target all incomplete tasks | - |\n| `N-M` | Task number range | - |\n| `--max-cycles N` | Maximum cycle count | `8` |\n| `--pacing \u003Cmode>` | Wake-up interval mode | `worker` (270s) |\n\n### Pacing Values\n\n| pacing | delaySeconds | Use case |\n|--------|-------------|----------|\n| `worker` | 270 | Immediately after Worker completion (within 5 min cache warm) |\n| `ci` | 270 | Waiting for short CI jobs |\n| `plateau` | 1200 | 20 min (retry interval after plateau detection) |\n| `night` | 3600 | Long overnight batch |\n\n> **Constraint**: `ScheduleWakeup`'s `delaySeconds` is clamped to **[60, 3600]** at runtime.\n> All pacing values are within this range. When specifying values directly, always use 60–3600.\n\n## Launch Flow (per wake-up entry)\n\nFull details: [`${CLAUDE_SKILL_DIR}\u002Freferences\u002Fflow.md`](${CLAUDE_SKILL_DIR}\u002Freferences\u002Fflow.md)\n\n```\nwake-up\n  │\n  ▼\n[Step 0] Flock-based concurrency guard\n  Prevent concurrent loop instances via lock directory\n  │\n  ▼\n[Step 0.5] State consistency check\n  bash tests\u002Fvalidate-plugin.sh --quick\n  │\n  ▼\n[Step 1] Read Plans.md first\n  Identify the leading cc:WIP \u002F cc:TODO task (get task_id)\n  No incomplete tasks → loop ends (normal completion)\n  │\n  ▼\n[Step 2] Check sprint-contract existence & generate\n  Check .claude\u002Fstate\u002Fcontracts\u002F${task_id}.sprint-contract.json\n  If absent: node \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fgenerate-sprint-contract.js\" ${task_id}\n  On first generation: bash \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fenrich-sprint-contract.sh\" \u003Ccontract-path> \\\n    --check \"auto-approve (harness-schedule-run — confirm DoD from reviewer perspective)\" \\\n    --approve  ← draft → approved\n  (Existing contracts already approved — skip)\n  │\n  ▼\n[Step 3] Contract readiness check\n  bash \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fensure-sprint-contract-ready.sh\" \u003Ccontract-path>\n  │\n  ▼\n[Step 4] Resume pack reload\n  harness-mem resume-pack (context re-injection)\n  │\n  ▼\n[Step 5] Execute 1 task cycle\n  worker_result = Agent(\n      subagent_type=\"harness:worker\",\n      prompt=\"Task: ${task_id}\\nDoD: \u003Cextracted from Plans.md>\\ncontract_path: ${CONTRACT_PATH}\\nmode: breezing\",\n      isolation=\"worktree\",\n      run_in_background=false\n  )\n  # worker_result: { commit, branch, worktreePath, files_changed, summary }\n  │\n  ▼\n[Step 5.5] Lead review execution\n  diff_text = git show worker_result.commit\n  verdict = codex_exec_review(diff_text) or reviewer_agent_review(diff_text)\n  See flow.md for details\n  │\n  ▼\n[Step 5.6] APPROVE → cherry-pick to main \u002F REQUEST_CHANGES → fix loop (max_iterations from contract, default 3)\n  APPROVE: git cherry-pick → update Plans.md to cc:Done [{hash}] → delete feature branch\n  REQUEST_CHANGES x MAX_REVIEWS still rejected: escalation\n  See flow.md for details\n  │\n  ▼\n[Step 6] Plateau detection\n  bash \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fdetect-review-plateau.sh\" ${current_task_id}\n  │\n  ├── PIVOT_REQUIRED (exit 2)   → loop stop + user escalation\n  ├── INSUFFICIENT_DATA (exit 1) → continue\n  └── PIVOT_NOT_REQUIRED (exit 0) → continue\n  │\n  ▼\n[Step 7] Cycle count check\n  │\n  ├── cycles >= max_cycles → loop stop (limit reached)\n  │\n  ▼\n[Step 8] Record checkpoint\n  harness_mem_record_checkpoint(\n      session_id, title, content=cycle result summary\n  )\n  │\n  ▼\n[Step 9] Schedule next wake-up\n  ScheduleWakeup(\n      delaySeconds=\u003Cpacing value>,\n      prompt=\"\u002Fharness-schedule-run \u003Csame args>\",\n      reason=\"Cycle {N}\u002F{max} complete — proceeding to next task\"\n  )\n```\n\n## Cycle Stop Conditions\n\n| Condition | Stop Type | Response |\n|-----------|-----------|----------|\n| `cycles >= max_cycles` | Normal stop (limit reached) | Report to user |\n| `PIVOT_REQUIRED` (exit 2) | Abnormal stop (escalation) | Ask user for decision |\n| No incomplete tasks | Normal stop (all complete) | Output completion report |\n\nWith `--max-cycles 3`, stops after 3 completed cycles.\nDefault (`--max-cycles 8`) stops after 8 cycles.\n\n## \u002Floop Integration\n\nThis skill is used in combination with CC's `\u002Floop` (dynamic mode).\n\nWhen `\u002Floop` is enabled, CC continues autonomous re-entry, scheduling the next wake-up via `ScheduleWakeup` at the end of each cycle.\n\n`\u002Floop` sentinel: `\u003C\u003Cautonomous-loop-dynamic>>`\n\nEach wake-up starts with a **fresh context**, preventing context contamination from the previous cycle.\nReload via `harness-mem resume-pack` is required (Step 4).\n\n## Checkpoint Schema\n\n```json\n{\n  \"session_id\": \"\u003Csession ID>\",\n  \"title\": \"harness-schedule-run cycle {N}\u002F{max}: {task name}\",\n  \"content\": \"1-line summary of cycle_result + commit hash\"\n}\n```\n\n## Related Skills\n\n- `harness-work` — Task implementation skill executed each cycle\n- `harness-plan` — Plan tasks targeted by the loop\n- `harness-review` — Review individual tasks\n- `session-control` — Session state management\n","# Harness Schedule Run\n\nMeta-skill that combines `\u002Floop` (CC dynamic mode) with `ScheduleWakeup` to re-enter long-running tasks with a **fresh context on every wake-up**.\n\nEach wake-up calls `harness-work --breezing` via the Agent tool, forming a re-entrant loop of 1 cycle = 1 task completion.\n\n## Quick Reference\n\n| Input | Behavior |\n|-------|----------|\n| `\u002Fharness-schedule-run all` | Loop all incomplete tasks (default: max 8 cycles) |\n| `\u002Fharness-schedule-run all --max-cycles 3` | Stop after 3 cycles |\n| `\u002Fharness-schedule-run 41.1-41.3 --pacing ci` | Execute task range with CI pacing |\n| `\u002Fharness-schedule-run all --pacing night` | Overnight batch (3600s interval) |\n\n## Options\n\n| Option | Description | Default |\n|--------|-------------|---------|\n| `all` | Target all incomplete tasks | - |\n| `N-M` | Task number range | - |\n| `--max-cycles N` | Maximum cycle count | `8` |\n| `--pacing \u003Cmode>` | Wake-up interval mode | `worker` (270s) |\n\n### Pacing Values\n\n| pacing | delaySeconds | Use case |\n|--------|-------------|----------|\n| `worker` | 270 | Immediately after Worker completion (within 5 min cache warm) |\n| `ci` | 270 | Waiting for short CI jobs |\n| `plateau` | 1200 | 20 min (retry interval after plateau detection) |\n| `night` | 3600 | Long overnight batch |\n\n> **Constraint**: `ScheduleWakeup`'s `delaySeconds` is clamped to **[60, 3600]** at runtime.\n> All pacing values are within this range. When specifying values directly, always use 60–3600.\n\n## Launch Flow (per wake-up entry)\n\nFull details: [`${CLAUDE_SKILL_DIR}\u002Freferences\u002Fflow.md`](${CLAUDE_SKILL_DIR}\u002Freferences\u002Fflow.md)\n\n```\nwake-up\n  │\n  ▼\n[Step 0] Flock-based concurrency guard\n  Prevent concurrent loop instances via lock directory\n  │\n  ▼\n[Step 0.5] State consistency check\n  bash tests\u002Fvalidate-plugin.sh --quick\n  │\n  ▼\n[Step 1] Read Plans.md first\n  Identify the leading cc:WIP \u002F cc:TODO task (get task_id)\n  No incomplete tasks → loop ends (normal completion)\n  │\n  ▼\n[Step 2] Check sprint-contract existence & generate\n  Check .claude\u002Fstate\u002Fcontracts\u002F${task_id}.sprint-contract.json\n  If absent: node \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fgenerate-sprint-contract.js\" ${task_id}\n  On first generation: bash \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fenrich-sprint-contract.sh\" \u003Ccontract-path> \\\n    --check \"auto-approve (harness-schedule-run — confirm DoD from reviewer perspective)\" \\\n    --approve  ← draft → approved\n  (Existing contracts already approved — skip)\n  │\n  ▼\n[Step 3] Contract readiness check\n  bash \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fensure-sprint-contract-ready.sh\" \u003Ccontract-path>\n  │\n  ▼\n[Step 4] Resume pack reload\n  harness-mem resume-pack (context re-injection)\n  │\n  ▼\n[Step 5] Execute 1 task cycle\n  worker_result = Agent(\n      subagent_type=\"harness:worker\",\n      prompt=\"Task: ${task_id}\\nDoD: \u003Cextracted from Plans.md>\\ncontract_path: ${CONTRACT_PATH}\\nmode: breezing\",\n      isolation=\"worktree\",\n      run_in_background=false\n  )\n  # worker_result: { commit, branch, worktreePath, files_changed, summary }\n  │\n  ▼\n[Step 5.5] Lead review execution\n  diff_text = git show worker_result.commit\n  verdict = codex_exec_review(diff_text) or reviewer_agent_review(diff_text)\n  See flow.md for details\n  │\n  ▼\n[Step 5.6] APPROVE → cherry-pick to main \u002F REQUEST_CHANGES → fix loop (max_iterations from contract, default 3)\n  APPROVE: git cherry-pick → update Plans.md to cc:Done [{hash}] → delete feature branch\n  REQUEST_CHANGES x MAX_REVIEWS still rejected: escalation\n  See flow.md for details\n  │\n  ▼\n[Step 6] Plateau detection\n  bash \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fdetect-review-plateau.sh\" ${current_task_id}\n  │\n  ├── PIVOT_REQUIRED (exit 2)   → loop stop + user escalation\n  ├── INSUFFICIENT_DATA (exit 1) → continue\n  └── PIVOT_NOT_REQUIRED (exit 0) → continue\n  │\n  ▼\n[Step 7] Cycle count check\n  │\n  ├── cycles >= max_cycles → loop stop (limit reached)\n  │\n  ▼\n[Step 8] Record checkpoint\n  harness_mem_record_checkpoint(\n      session_id, title, content=cycle result summary\n  )\n  │\n  ▼\n[Step 9] Schedule next wake-up\n  ScheduleWakeup(\n      delaySeconds=\u003Cpacing value>,\n      prompt=\"\u002Fharness-schedule-run \u003Csame args>\",\n      reason=\"Cycle {N}\u002F{max} complete — proceeding to next task\"\n  )\n```\n\n## Cycle Stop Conditions\n\n| Condition | Stop Type | Response |\n|-----------|-----------|----------|\n| `cycles >= max_cycles` | Normal stop (limit reached) | Report to user |\n| `PIVOT_REQUIRED` (exit 2) | Abnormal stop (escalation) | Ask user for decision |\n| No incomplete tasks | Normal stop (all complete) | Output completion report |\n\nWith `--max-cycles 3`, stops after 3 completed cycles.\nDefault (`--max-cycles 8`) stops after 8 cycles.\n\n## \u002Floop Integration\n\nThis skill is used in combination with CC's `\u002Floop` (dynamic mode).\n\nWhen `\u002Floop` is enabled, CC continues autonomous re-entry, scheduling the next wake-up via `ScheduleWakeup` at the end of each cycle.\n\n`\u002Floop` sentinel: `\u003C\u003Cautonomous-loop-dynamic>>`\n\nEach wake-up starts with a **fresh context**, preventing context contamination from the previous cycle.\nReload via `harness-mem resume-pack` is required (Step 4).\n\n## Checkpoint Schema\n\n```json\n{\n  \"session_id\": \"\u003Csession ID>\",\n  \"title\": \"harness-schedule-run cycle {N}\u002F{max}: {task name}\",\n  \"content\": \"1-line summary of cycle_result + commit hash\"\n}\n```\n\n## Related Skills\n\n- `harness-work` — Task implementation skill executed each cycle\n- `harness-plan` — Plan tasks targeted by the loop\n- `harness-review` — Review individual tasks\n- `session-control` — Session state management\n",{"name":5,"description":7,"allowed-tools":49,"argument-hint":56},[50,51,40,52,53,54,55],"Read","Edit","Task","ScheduleWakeup","mcp__harness__harness_mem_resume_pack","mcp__harness__harness_mem_record_checkpoint","[all|N-M] [--max-cycles N] [--pacing worker|ci|plateau|night]","master","cf086c6a667d37610d30a4236b94d8ab02e4efc5","harness\u002Ftemplates\u002Fcodex-skills\u002Fharness-schedule-run",0,"https:\u002F\u002Fgithub.com\u002Ftim-hub\u002Fpowerball-harness","https:\u002F\u002Fgithub.com\u002Ftim-hub","https:\u002F\u002Ftim.bai.uno","Building things.","2026-08-05 17:45:08",{"source":8,"skillName":5,"commands":67,"prompt":77,"zipAvailable":78},[68,71,74],{"tool":69,"command":70},"npx","npx skills add tim-hub\u002Fpowerball-harness --skill harness-schedule-run",{"tool":72,"command":73},"bunx","bunx skills add tim-hub\u002Fpowerball-harness --skill harness-schedule-run",{"tool":75,"command":76},"pnpm","pnpm dlx skills add tim-hub\u002Fpowerball-harness --skill harness-schedule-run","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Ftim-hub\u002Fpowerball-harness\u002Fblob\u002Fmaster\u002Fharness\u002Ftemplates\u002Fcodex-skills\u002Fharness-schedule-run\u002FSKILL.md",true,[80],{"path":81,"resourceType":82,"resourceTypeDesc":83,"sizeBytes":84,"sourceHtmlUrl":85},"references\u002Fflow.md","REFERENCE","参考文档",21019,"https:\u002F\u002Fgithub.com\u002Ftim-hub\u002Fpowerball-harness\u002Fblob\u002Fmaster\u002Fharness\u002Ftemplates\u002Fcodex-skills\u002Fharness-schedule-run\u002Freferences\u002Fflow.md",[87,101,109,115,135,147],{"skillNo":88,"name":89,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":93,"publishedAt":44},"01M6R7GH70ABV40CS95W4WWMQZ","escape-skill","https:\u002F\u002Fgithub.com\u002Ftim-hub\u002Fpowerball-harness\u002Fblob\u002Fmaster\u002Ftests\u002Ffixtures\u002Fskills\u002Fzz-review-escape\u002FSKILL.md",[92],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[94,95,96,97,98,99,100],"escape","escape skill","quoted","path","true","false","Tools",{"skillNo":102,"name":103,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":106,"publishedAt":44},"01M6R7G8A2BQHYTAR39BY7X3GK","zz-review-empty","https:\u002F\u002Fgithub.com\u002Ftim-hub\u002Fpowerball-harness\u002Fblob\u002Fmaster\u002Ftests\u002Ffixtures\u002Fskills\u002Fzz-review-empty\u002FSKILL.md",[],[43,107,108,98,99],"zz review empty","empty",{"skillNo":110,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":111,"categories":112,"keywords":114,"publishedAt":44},"01M6R7G3XH6C43JKP2RX1SZW6V","https:\u002F\u002Fgithub.com\u002Ftim-hub\u002Fpowerball-harness\u002Fblob\u002Fmaster\u002Fharness\u002Ftemplates\u002Fopencode\u002Fskills\u002Fharness-schedule-run\u002FSKILL.md",[113],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[26,27,25,24,42,43,37,38,31,29,30,41,32,33,34,35,36,28,39,40],{"skillNo":116,"name":117,"nameZh":6,"description":118,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":119,"categories":120,"keywords":121,"publishedAt":44},"01M6R7FY3ZVYDDKXBT3D7QSFFK","harness-work","Use when implementing, executing, or running Plans.md tasks — single task, parallel workers, or full team\u002Fbreezing run. Accepts specific task numbers or ranges. Do NOT load for: planning, review, release, or setup.","https:\u002F\u002Fgithub.com\u002Ftim-hub\u002Fpowerball-harness\u002Fblob\u002Fmaster\u002Fharness\u002Ftemplates\u002Fcodex-skills\u002Fharness-work\u002FSKILL.md",[],[122,39,26,123,124,29,125,126,127,128,129,130,131,132,133,25,42,134,43,40],"harness work","implementing","read","breezing","mode","number","full","write","edit","auto","team","release","commit",{"skillNo":136,"name":125,"nameZh":6,"description":137,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":138,"categories":139,"keywords":140,"publishedAt":44},"01M6R7FCDN9AS7RSQFGZWHTF7F","Use when running the full team\u002Fbreezing flow end-to-end — all tasks with parallel workers. Do NOT load for: single-task implementation, planning, review, release, or setup.","https:\u002F\u002Fgithub.com\u002Ftim-hub\u002Fpowerball-harness\u002Fblob\u002Fmaster\u002Fharness\u002Ftemplates\u002Fcodex-skills\u002Fbreezing\u002FSKILL.md",[],[125,42,134,43,133,141,142,29,143,126,144,128,131,130,129,124,145,40,132,146],"websearch","running","single-task","end","parallel","flow",{"skillNo":148,"name":149,"nameZh":6,"description":150,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":151,"categories":152,"keywords":153,"publishedAt":44},"01M6R7F867HKSBNS7DDQ09ZE8H","maintenance","Periodic housekeeping and session lifecycle management. Use when performing cleanup, pruning, or session commands.","https:\u002F\u002Fgithub.com\u002Ftim-hub\u002Fpowerball-harness\u002Fblob\u002Fmaster\u002Fharness\u002Fskills\u002Fmaintenance\u002FSKILL.md",[],[149,154,29,98,155,156,157,158,159,130,129,124,160,40,161,162,163,164,165,166],"session","message","clean","worktrees","periodic","use","check","lifecycle","cache","inbox","state","cleanup","sessions"]