[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M75C1GPAN6T32G578ZCQTYZ4":3,"skill-same:01M75C1GPAN6T32G578ZCQTYZ4":48},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":30,"ownerHtmlUrl":31,"ownerBlogUrl":32,"ownerDescription":6,"updatedAt":33,"install":34,"resources":47},"01M75C1GPAN6T32G578ZCQTYZ4","cancel",null,"Cancel any active OMC mode (autopilot, ralph, ultrawork, ultraqa, swarm, ultrapilot, pipeline, team)","agisota\u002Fold-one","agisota","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F121413592?v=4","01M75B4BRNT67EC1NDRT8QZEP2","01M75B4BS7QR9VP580MSCPTS3X",3,"MIT","https:\u002F\u002Fgithub.com\u002Fagisota\u002Fold-one\u002Fblob\u002Fmain\u002Fskills\u002Fcancel\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: cancel\ndescription: Cancel any active OMC mode (autopilot, ralph, ultrawork, ultraqa, swarm, ultrapilot, pipeline, team)\nargument-hint: \"[--force|--all]\"\nlevel: 2\n---\n\n# Cancel Skill\n\nIntelligent cancellation that detects and cancels the active OMC mode.\n\n**The cancel skill is the standard way to complete and exit any OMC mode.**\nWhen the stop hook detects work is complete, it instructs the LLM to invoke\nthis skill for proper state cleanup. If cancel fails or is interrupted,\nretry with `--force` flag, or wait for the 2-hour staleness timeout as\na last resort.\n\n## What It Does\n\nAutomatically detects which mode is active and cancels it:\n- **Autopilot**: Stops workflow, preserves progress for resume\n- **Ralph**: Stops persistence loop, clears linked ultrawork if applicable\n- **Ultrawork**: Stops parallel execution (standalone or linked)\n- **UltraQA**: Stops QA cycling workflow\n- **Swarm**: Stops coordinated agent swarm, releases claimed tasks\n- **Ultrapilot**: Stops parallel autopilot workers\n- **Pipeline**: Stops sequential agent pipeline\n- **Team**: Sends shutdown_request to all teammates, waits for responses, calls TeamDelete, clears linked ralph if present\n- **Team+Ralph (linked)**: Cancels team first (graceful shutdown), then clears ralph state. Cancelling ralph when linked also cancels team first.\n\n## Usage\n\n```\n\u002Foh-my-claudecode:cancel\n```\n\nOr say: \"cancelomc\", \"stopomc\"\n\n## Critical: Deferred Tool Handling\n\nThe state management tools (`state_clear`, `state_read`, `state_write`, `state_list_active`,\n`state_get_status`) may be registered as **deferred tools** by Claude Code. Before calling\nany state tool, you MUST first load all of them via `ToolSearch`:\n\n```\nToolSearch(query=\"select:mcp__plugin_oh-my-claudecode_t__state_clear,mcp__plugin_oh-my-claudecode_t__state_read,mcp__plugin_oh-my-claudecode_t__state_write,mcp__plugin_oh-my-claudecode_t__state_list_active,mcp__plugin_oh-my-claudecode_t__state_get_status\")\n```\n\nIf `state_clear` is unavailable or fails, use this **bash fallback** as an **emergency\nescape from the stop hook loop**. This is NOT a full replacement for the cancel flow —\nit only removes state files to unblock the session. Linked modes (e.g. ralph→ultrawork,\nautopilot→ralph\u002Fultraqa) must be cleared separately by running the fallback once per mode.\n\nReplace `MODE` with the specific mode (e.g. `ralplan`, `ralph`, `ultrawork`, `ultraqa`).\n\n**WARNING:** Do NOT use this fallback for `autopilot` or `omc-teams`. Autopilot requires\n`state_write(active=false)` to preserve resume data. omc-teams requires tmux session\ncleanup that cannot be done via file deletion alone.\n\n```bash\n# Fallback: direct file removal when state_clear MCP tool is unavailable\nSESSION_ID=\"${CLAUDE_SESSION_ID:-${CLAUDECODE_SESSION_ID:-}}\"\nREPO_ROOT=\"$(git rev-parse --show-toplevel 2>\u002Fdev\u002Fnull || { d=\"$PWD\"; while [ \"$d\" != \"\u002F\" ] && [ ! -d \"$d\u002F.omc\" ]; do d=\"$(dirname \"$d\")\"; done; echo \"$d\"; })\"\n\n# Cross-platform SHA-256 (macOS: shasum, Linux: sha256sum)\nsha256portable() { printf '%s' \"$1\" | (sha256sum 2>\u002Fdev\u002Fnull || shasum -a 256) | cut -c1-16; }\n\n# Resolve state directory (supports OMC_STATE_DIR centralized storage)\nif [ -n \"${OMC_STATE_DIR:-}\" ]; then\n  # Mirror getProjectIdentifier() from worktree-paths.ts\n  SOURCE=\"$(git remote get-url origin 2>\u002Fdev\u002Fnull || echo \"$REPO_ROOT\")\"\n  HASH=\"$(sha256portable \"$SOURCE\")\"\n  DIR_NAME=\"$(basename \"$REPO_ROOT\" | sed 's\u002F[^a-zA-Z0-9_-]\u002F_\u002Fg')\"\n  OMC_STATE=\"$OMC_STATE_DIR\u002F${DIR_NAME}-${HASH}\u002Fstate\"\n  [ ! -d \"$OMC_STATE\" ] && { echo \"ERROR: State dir not found at $OMC_STATE\" >&2; exit 1; }\nelif [ \"$REPO_ROOT\" != \"\u002F\" ] && [ -d \"$REPO_ROOT\u002F.omc\" ]; then\n  OMC_STATE=\"$REPO_ROOT\u002F.omc\u002Fstate\"\nelse\n  echo \"ERROR: Could not locate .omc state directory\" >&2\n  exit 1\nfi\nMODE=\"ralplan\"  # \u003C-- replace with the target mode\n\n# Clear session-scoped state for the specific mode\nif [ -n \"$SESSION_ID\" ] && [ -d \"$OMC_STATE\u002Fsessions\u002F$SESSION_ID\" ]; then\n  rm -f \"$OMC_STATE\u002Fsessions\u002F$SESSION_ID\u002F${MODE}-state.json\"\n  rm -f \"$OMC_STATE\u002Fsessions\u002F$SESSION_ID\u002F${MODE}-stop-breaker.json\"\n  rm -f \"$OMC_STATE\u002Fsessions\u002F$SESSION_ID\u002Fskill-active-state.json\"\n  # Write cancel signal so stop hook detects cancellation in progress\n  NOW_ISO=\"$(date -u +\"%Y-%m-%dT%H:%M:%SZ\")\"\n  EXPIRES_ISO=\"$(date -u -d \"+30 seconds\" +\"%Y-%m-%dT%H:%M:%SZ\" 2>\u002Fdev\u002Fnull || python3 - \u003C\u003C'PY'\\nfrom datetime import datetime, timedelta, timezone\\nprint((datetime.now(timezone.utc) + timedelta(seconds=30)).strftime('%Y-%m-%dT%H:%M:%SZ'))\\nPY\\n)\"\n  printf '{\"active\":true,\"requested_at\":\"%s\",\"expires_at\":\"%s\",\"mode\":\"%s\",\"source\":\"bash_fallback\"}' \\\n    \"$NOW_ISO\" \"$EXPIRES_ISO\" \"$MODE\" > \"$OMC_STATE\u002Fsessions\u002F$SESSION_ID\u002Fcancel-signal-state.json\"\nfi\n\n# Clear legacy state only if no session ID (avoid clearing another session's state)\nif [ -z \"$SESSION_ID\" ]; then\n  rm -f \"$OMC_STATE\u002F${MODE}-state.json\"\nfi\n```\n\n## Auto-Detection\n\n`\u002Foh-my-claudecode:cancel` follows the session-aware state contract:\n- By default the command inspects the current session via `state_list_active` and `state_get_status`, navigating `.omc\u002Fstate\u002Fsessions\u002F{sessionId}\u002F…` to discover which mode is active.\n- When a session id is provided or already known, that session-scoped path is authoritative. Legacy files in `.omc\u002Fstate\u002F*.json` are consulted only as a compatibility fallback if the session id is missing or empty.\n- Swarm is a shared SQLite\u002Fmarker mode (`.omc\u002Fstate\u002Fswarm.db` \u002F `.omc\u002Fstate\u002Fswarm-active.marker`) and is not session-scoped.\n- The default cleanup flow calls `state_clear` with the session id to remove only the matching session files; modes stay bound to their originating session.\n\nActive modes are still cancelled in dependency order:\n1. Autopilot (includes linked ralph\u002Fultraqa\u002F cleanup)\n2. Ralph (cleans its linked ultrawork or )\n3. Ultrawork (standalone)\n4. UltraQA (standalone)\n5. Swarm (standalone)\n6. Ultrapilot (standalone)\n7. Pipeline (standalone)\n8. Team (Claude Code native)\n9. OMC Teams (tmux CLI workers)\n10. Plan Consensus (standalone)\n11. Self-Improve (standalone — clear state, clean orphaned worktrees, preserve iteration_state for resume, set status: \"user_stopped\" in the resolved `\u003Cself-improve-root>\u002Fstate\u002Fagent-settings.json`; new runs use `.omc\u002Fself-improve\u002Ftopics\u002F\u003Ctopic-slug>\u002F`, with flat `.omc\u002Fself-improve\u002F` retained only for legacy single-track resumes)\n\n## Force Clear All\n\nUse `--force` or `--all` when you need to erase every session plus legacy artifacts, e.g., to reset the workspace entirely.\n\n```\n\u002Foh-my-claudecode:cancel --force\n```\n\n```\n\u002Foh-my-claudecode:cancel --all\n```\n\nSteps under the hood:\n1. `state_list_active` enumerates `.omc\u002Fstate\u002Fsessions\u002F{sessionId}\u002F…` to find every known session.\n2. `state_clear` runs once per session to drop that session’s files.\n3. A global `state_clear` without `session_id` removes legacy files under `.omc\u002Fstate\u002F*.json`, `.omc\u002Fstate\u002Fswarm*.db`, and compatibility artifacts (see list).\n4. Team artifacts (`~\u002F.claude\u002Fteams\u002F*\u002F`, `~\u002F.claude\u002Ftasks\u002F*\u002F`, `.omc\u002Fstate\u002Fteam-state.json`) are best-effort cleared as part of the legacy fallback.\n   - Cancel for native team does NOT affect omc-teams state, and vice versa.\n\nEvery `state_clear` command honors the `session_id` argument, so even force mode still uses the session-aware paths first before deleting legacy files.\n\nLegacy compatibility list (removed only under `--force`\u002F`--all`):\n- `.omc\u002Fstate\u002Fautopilot-state.json`\n- `.omc\u002Fstate\u002Fralph-state.json`\n- `.omc\u002Fstate\u002Fralph-plan-state.json`\n- `.omc\u002Fstate\u002Fralph-verification.json`\n- `.omc\u002Fstate\u002Fultrawork-state.json`\n- `.omc\u002Fstate\u002Fultraqa-state.json`\n- `.omc\u002Fstate\u002Fswarm.db`\n- `.omc\u002Fstate\u002Fswarm.db-wal`\n- `.omc\u002Fstate\u002Fswarm.db-shm`\n- `.omc\u002Fstate\u002Fswarm-active.marker`\n- `.omc\u002Fstate\u002Fswarm-tasks.db`\n- `.omc\u002Fstate\u002Fultrapilot-state.json`\n- `.omc\u002Fstate\u002Fultrapilot-ownership.json`\n- `.omc\u002Fstate\u002Fpipeline-state.json`\n- `.omc\u002Fstate\u002Fomc-teams-state.json`\n- `.omc\u002Fstate\u002Fplan-consensus.json`\n- `.omc\u002Fstate\u002Fralplan-state.json`\n- `.omc\u002Fstate\u002Fboulder.json`\n- `.omc\u002Fstate\u002Fhud-state.json`\n- `.omc\u002Fstate\u002Fsubagent-tracking.json`\n- `.omc\u002Fstate\u002Fsubagent-tracker.lock`\n- `.omc\u002Fstate\u002Frate-limit-daemon.pid`\n- `.omc\u002Fstate\u002Frate-limit-daemon.log`\n- `.omc\u002Fstate\u002Fcheckpoints\u002F` (directory)\n- `.omc\u002Fstate\u002Fsessions\u002F` (empty directory cleanup after clearing sessions)\n\n## Implementation Steps\n\nWhen you invoke this skill:\n\n### 1. Parse Arguments\n\n```bash\n# Check for --force or --all flags\nFORCE_MODE=false\nif [[ \"$*\" == *\"--force\"* ]] || [[ \"$*\" == *\"--all\"* ]]; then\n  FORCE_MODE=true\nfi\n```\n\n### 2. Detect Active Modes\n\nThe skill now relies on the session-aware state contract rather than hard-coded file paths:\n1. Call `state_list_active` to enumerate `.omc\u002Fstate\u002Fsessions\u002F{sessionId}\u002F…` and discover every active session.\n2. For each session id, call `state_get_status` to learn which mode is running (`autopilot`, `ralph`, `ultrawork`, etc.) and whether dependent modes exist.\n3. If a `session_id` was supplied to `\u002Foh-my-claudecode:cancel`, skip legacy fallback entirely and operate solely within that session path; otherwise, consult legacy files in `.omc\u002Fstate\u002F*.json` only if the state tools report no active session. Swarm remains a shared SQLite\u002Fmarker mode outside session scoping.\n4. Any cancellation logic in this doc mirrors the dependency order discovered via state tools (autopilot → ralph → …).\n\n### 3A. Force Mode (if --force or --all)\n\nUse force mode to clear every session plus legacy artifacts via `state_clear`. Direct file removal is reserved for legacy cleanup when the state tools report no active sessions.\n\n### 3B. Smart Cancellation (default)\n\n#### If Team Active (Claude Code native)\n\nTeams are detected by checking for config files in `${CLAUDE_CONFIG_DIR:-~\u002F.claude}\u002Fteams\u002F`:\n\n```bash\n# Check for active teams\nTEAM_CONFIGS=$(find \"${CLAUDE_CONFIG_DIR:-$HOME\u002F.claude}\"\u002Fteams -name config.json -maxdepth 2 2>\u002Fdev\u002Fnull)\n```\n\n**Two-pass cancellation protocol:**\n\n**Pass 1: Graceful Shutdown**\n```\nFor each team found in ${CLAUDE_CONFIG_DIR:-~\u002F.claude}\u002Fteams\u002F:\n  1. Read config.json to get team_name and members list\n  2. For each non-lead member:\n     a. Send shutdown_request via SendMessage\n     b. Wait up to 15 seconds for shutdown_response\n     c. If response received: member terminates and is auto-removed\n     d. If timeout: mark member as unresponsive, continue to next\n  3. Log: \"Graceful pass: X\u002FY members responded\"\n```\n\n**Pass 2: Reconciliation**\n```\nAfter graceful pass:\n  1. Re-read config.json to check remaining members\n  2. If only lead remains (or config is empty): proceed to TeamDelete\n  3. If unresponsive members remain:\n     a. Wait 5 more seconds (they may still be processing)\n     b. Re-read config.json again\n     c. If still stuck: attempt TeamDelete anyway\n     d. If TeamDelete fails: report manual cleanup path\n```\n\n**TeamDelete + Cleanup:**\n```\n  1. Call TeamDelete() — removes ~\u002F.claude\u002Fteams\u002F{name}\u002F and ~\u002F.claude\u002Ftasks\u002F{name}\u002F\n  2. Clear team state: state_clear(mode=\"team\")\n  3. Check for linked ralph: state_read(mode=\"ralph\") — if linked_team is true:\n     a. Clear ralph state: state_clear(mode=\"ralph\")\n     b. Clear linked ultrawork if present: state_clear(mode=\"ultrawork\")\n  4. Run orphan scan (see below)\n  5. Emit structured cancel report\n```\n\n**Orphan Detection (Post-Cleanup):**\n\nAfter TeamDelete, verify no agent processes remain:\n```bash\nnode \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fcleanup-orphans.mjs\" --team-name \"{team_name}\"\n```\n\nThe orphan scanner:\n1. Checks `ps aux` (Unix) or `tasklist` (Windows) for processes with `--team-name` matching the deleted team\n2. For each orphan whose team config no longer exists: sends SIGTERM, waits 5s, sends SIGKILL if still alive\n3. Reports cleanup results as JSON\n\nUse `--dry-run` to inspect without killing. The scanner is safe to run multiple times.\n\n**Structured Cancel Report:**\n```\nTeam \"{team_name}\" cancelled:\n  - Members signaled: N\n  - Responses received: M\n  - Unresponsive: K (list names if any)\n  - TeamDelete: success\u002Ffailed\n  - Manual cleanup needed: yes\u002Fno\n    Path: ~\u002F.claude\u002Fteams\u002F{name}\u002F and ~\u002F.claude\u002Ftasks\u002F{name}\u002F\n```\n\n**Implementation note:** The cancel skill is executed by the LLM, not as a bash script. When you detect an active team:\n1. Read `${CLAUDE_CONFIG_DIR:-~\u002F.claude}\u002Fteams\u002F*\u002Fconfig.json` to find active teams\n2. If multiple teams exist, cancel oldest first (by `createdAt`)\n3. For each non-lead member, call `SendMessage(type: \"shutdown_request\", recipient: member-name, content: \"Cancelling\")`\n4. Wait briefly for shutdown responses (15s per member timeout)\n5. Re-read config.json to check for remaining members (reconciliation pass)\n6. Call `TeamDelete()` to clean up\n7. Clear team state: `state_clear(mode=\"team\", session_id)`\n8. Report structured summary to user\n\n#### If Autopilot Active\n\nAutopilot handles its own cleanup including linked ralph and ultraqa.\n\n1. Read autopilot state via `state_read(mode=\"autopilot\", session_id)` to get current phase\n2. Check for linked ralph via `state_read(mode=\"ralph\", session_id)`:\n   - If ralph is active and has `linked_ultrawork: true`, clear ultrawork first: `state_clear(mode=\"ultrawork\", session_id)`\n   - Clear ralph: `state_clear(mode=\"ralph\", session_id)`\n3. Check for linked ultraqa via `state_read(mode=\"ultraqa\", session_id)`:\n   - If active, clear it: `state_clear(mode=\"ultraqa\", session_id)`\n4. Mark autopilot inactive (preserve state for resume) via `state_write(mode=\"autopilot\", session_id, state={active: false, ...existing})`\n\n#### If Ralph Active (but not Autopilot)\n\n1. Read ralph state via `state_read(mode=\"ralph\", session_id)` to check for linked ultrawork\n2. If `linked_ultrawork: true`:\n   - Read ultrawork state to verify `linked_to_ralph: true`\n   - If linked, clear ultrawork: `state_clear(mode=\"ultrawork\", session_id)`\n3. Clear ralph: `state_clear(mode=\"ralph\", session_id)`\n\n#### If Ultrawork Active (standalone, not linked)\n\n1. Read ultrawork state via `state_read(mode=\"ultrawork\", session_id)`\n2. If `linked_to_ralph: true`, warn user to cancel ralph instead (which cascades)\n3. Otherwise clear: `state_clear(mode=\"ultrawork\", session_id)`\n\n#### If UltraQA Active (standalone)\n\nClear directly: `state_clear(mode=\"ultraqa\", session_id)`\n\n#### No Active Modes\n\nReport: \"No active OMC modes detected. Use --force to clear all state files anyway.\"\n\n## Implementation Notes\n\nThe cancel skill runs as follows:\n1. Parse the `--force` \u002F `--all` flags, tracking whether cleanup should span every session or stay scoped to the current session id.\n2. Use `state_list_active` to enumerate known session ids and `state_get_status` to learn the active mode (`autopilot`, `ralph`, `ultrawork`, etc.) for each session.\n3. When operating in default mode, call `state_clear` with that session_id to remove only the session’s files, then run mode-specific cleanup (autopilot → ralph → …) based on the state tool signals.\n4. In force mode, iterate every active session, call `state_clear` per session, then run a global `state_clear` without `session_id` to drop legacy files (`.omc\u002Fstate\u002F*.json`, compatibility artifacts) and report success. Swarm remains a shared SQLite\u002Fmarker mode outside session scoping.\n5. Team artifacts (`~\u002F.claude\u002Fteams\u002F*\u002F`, `~\u002F.claude\u002Ftasks\u002F*\u002F`, `.omc\u002Fstate\u002Fteam-state.json`) remain best-effort cleanup items invoked during the legacy\u002Fglobal pass.\n6. **Always** clear skill-active state as the final step, regardless of which mode was active or whether `--force` was used:\n   ```\n   state_clear(mode=\"skill-active\", session_id)\n   ```\n   This ensures the stop hook does not keep firing skill-protection reinforcements after cancel due to a stale `skill-active-state.json`. See issue #2118.\n\nState tools always honor the `session_id` argument, so even force mode still clears the session-scoped paths before deleting compatibility-only legacy state.\n\nMode-specific subsections below describe what extra cleanup each handler performs after the state-wide operations finish.\n## Messages Reference\n\n| Mode | Success Message |\n|------|-----------------|\n| Autopilot | \"Autopilot cancelled at phase: {phase}. Progress preserved for resume.\" |\n| Ralph | \"Ralph cancelled. Persistent mode deactivated.\" |\n| Ultrawork | \"Ultrawork cancelled. Parallel execution mode deactivated.\" |\n| UltraQA | \"UltraQA cancelled. QA cycling workflow stopped.\" |\n| Swarm | \"Swarm cancelled. Coordinated agents stopped.\" |\n| Ultrapilot | \"Ultrapilot cancelled. Parallel autopilot workers stopped.\" |\n| Pipeline | \"Pipeline cancelled. Sequential agent chain stopped.\" |\n| Team | \"Team cancelled. Teammates shut down and cleaned up.\" |\n| Plan Consensus | \"Plan Consensus cancelled. Planning session ended.\" |\n| Force | \"All OMC modes cleared. You are free to start fresh.\" |\n| None | \"No active OMC modes detected.\" |\n\n## What Gets Preserved\n\n| Mode | State Preserved | Resume Command |\n|------|-----------------|----------------|\n| Autopilot | Yes (phase, files, spec, plan, verdicts) | `\u002Foh-my-claudecode:autopilot` |\n| Ralph | No | N\u002FA |\n| Ultrawork | No | N\u002FA |\n| UltraQA | No | N\u002FA |\n| Swarm | No | N\u002FA |\n| Ultrapilot | No | N\u002FA |\n| Pipeline | No | N\u002FA |\n| Plan Consensus | Yes (plan file path preserved) | N\u002FA |\n\n## Notes\n\n- **Dependency-aware**: Autopilot cancellation cleans up Ralph and UltraQA\n- **Link-aware**: Ralph cancellation cleans up linked Ultrawork\n- **Safe**: Only clears linked Ultrawork, preserves standalone Ultrawork\n- **Local-only**: Clears state files in `.omc\u002Fstate\u002F` directory\n- **Resume-friendly**: Autopilot state is preserved for seamless resume\n- **Team-aware**: Detects native Claude Code teams and performs graceful shutdown\n\n## MCP Worker Cleanup\n\nWhen cancelling modes that may have spawned MCP workers (team bridge daemons), the cancel skill should also:\n\n1. **Check for active MCP workers**: Look for heartbeat files at `.omc\u002Fstate\u002Fteam-bridge\u002F{team}\u002F*.heartbeat.json`\n2. **Send shutdown signals**: Write shutdown signal files for each active worker\n3. **Kill tmux sessions**: Run `tmux kill-session -t omc-team-{team}-{worker}` for each worker\n4. **Clean up heartbeat files**: Remove all heartbeat files for the team\n5. **Clean up shadow registry**: Remove `.omc\u002Fstate\u002Fteam-mcp-workers.json`\n\n### Force Clear Addition\n\nWhen `--force` is used, also clean up:\n```bash\nrm -rf .omc\u002Fstate\u002Fteam-bridge\u002F       # Heartbeat files\nrm -f .omc\u002Fstate\u002Fteam-mcp-workers.json  # Shadow registry\n# Kill all omc-team-* tmux sessions\ntmux list-sessions -F '#{session_name}' 2>\u002Fdev\u002Fnull | grep '^omc-team-' | while read s; do tmux kill-session -t \"$s\" 2>\u002Fdev\u002Fnull; done\n```\n","# Cancel Skill\n\nIntelligent cancellation that detects and cancels the active OMC mode.\n\n**The cancel skill is the standard way to complete and exit any OMC mode.**\nWhen the stop hook detects work is complete, it instructs the LLM to invoke\nthis skill for proper state cleanup. If cancel fails or is interrupted,\nretry with `--force` flag, or wait for the 2-hour staleness timeout as\na last resort.\n\n## What It Does\n\nAutomatically detects which mode is active and cancels it:\n- **Autopilot**: Stops workflow, preserves progress for resume\n- **Ralph**: Stops persistence loop, clears linked ultrawork if applicable\n- **Ultrawork**: Stops parallel execution (standalone or linked)\n- **UltraQA**: Stops QA cycling workflow\n- **Swarm**: Stops coordinated agent swarm, releases claimed tasks\n- **Ultrapilot**: Stops parallel autopilot workers\n- **Pipeline**: Stops sequential agent pipeline\n- **Team**: Sends shutdown_request to all teammates, waits for responses, calls TeamDelete, clears linked ralph if present\n- **Team+Ralph (linked)**: Cancels team first (graceful shutdown), then clears ralph state. Cancelling ralph when linked also cancels team first.\n\n## Usage\n\n```\n\u002Foh-my-claudecode:cancel\n```\n\nOr say: \"cancelomc\", \"stopomc\"\n\n## Critical: Deferred Tool Handling\n\nThe state management tools (`state_clear`, `state_read`, `state_write`, `state_list_active`,\n`state_get_status`) may be registered as **deferred tools** by Claude Code. Before calling\nany state tool, you MUST first load all of them via `ToolSearch`:\n\n```\nToolSearch(query=\"select:mcp__plugin_oh-my-claudecode_t__state_clear,mcp__plugin_oh-my-claudecode_t__state_read,mcp__plugin_oh-my-claudecode_t__state_write,mcp__plugin_oh-my-claudecode_t__state_list_active,mcp__plugin_oh-my-claudecode_t__state_get_status\")\n```\n\nIf `state_clear` is unavailable or fails, use this **bash fallback** as an **emergency\nescape from the stop hook loop**. This is NOT a full replacement for the cancel flow —\nit only removes state files to unblock the session. Linked modes (e.g. ralph→ultrawork,\nautopilot→ralph\u002Fultraqa) must be cleared separately by running the fallback once per mode.\n\nReplace `MODE` with the specific mode (e.g. `ralplan`, `ralph`, `ultrawork`, `ultraqa`).\n\n**WARNING:** Do NOT use this fallback for `autopilot` or `omc-teams`. Autopilot requires\n`state_write(active=false)` to preserve resume data. omc-teams requires tmux session\ncleanup that cannot be done via file deletion alone.\n\n```bash\n# Fallback: direct file removal when state_clear MCP tool is unavailable\nSESSION_ID=\"${CLAUDE_SESSION_ID:-${CLAUDECODE_SESSION_ID:-}}\"\nREPO_ROOT=\"$(git rev-parse --show-toplevel 2>\u002Fdev\u002Fnull || { d=\"$PWD\"; while [ \"$d\" != \"\u002F\" ] && [ ! -d \"$d\u002F.omc\" ]; do d=\"$(dirname \"$d\")\"; done; echo \"$d\"; })\"\n\n# Cross-platform SHA-256 (macOS: shasum, Linux: sha256sum)\nsha256portable() { printf '%s' \"$1\" | (sha256sum 2>\u002Fdev\u002Fnull || shasum -a 256) | cut -c1-16; }\n\n# Resolve state directory (supports OMC_STATE_DIR centralized storage)\nif [ -n \"${OMC_STATE_DIR:-}\" ]; then\n  # Mirror getProjectIdentifier() from worktree-paths.ts\n  SOURCE=\"$(git remote get-url origin 2>\u002Fdev\u002Fnull || echo \"$REPO_ROOT\")\"\n  HASH=\"$(sha256portable \"$SOURCE\")\"\n  DIR_NAME=\"$(basename \"$REPO_ROOT\" | sed 's\u002F[^a-zA-Z0-9_-]\u002F_\u002Fg')\"\n  OMC_STATE=\"$OMC_STATE_DIR\u002F${DIR_NAME}-${HASH}\u002Fstate\"\n  [ ! -d \"$OMC_STATE\" ] && { echo \"ERROR: State dir not found at $OMC_STATE\" >&2; exit 1; }\nelif [ \"$REPO_ROOT\" != \"\u002F\" ] && [ -d \"$REPO_ROOT\u002F.omc\" ]; then\n  OMC_STATE=\"$REPO_ROOT\u002F.omc\u002Fstate\"\nelse\n  echo \"ERROR: Could not locate .omc state directory\" >&2\n  exit 1\nfi\nMODE=\"ralplan\"  # \u003C-- replace with the target mode\n\n# Clear session-scoped state for the specific mode\nif [ -n \"$SESSION_ID\" ] && [ -d \"$OMC_STATE\u002Fsessions\u002F$SESSION_ID\" ]; then\n  rm -f \"$OMC_STATE\u002Fsessions\u002F$SESSION_ID\u002F${MODE}-state.json\"\n  rm -f \"$OMC_STATE\u002Fsessions\u002F$SESSION_ID\u002F${MODE}-stop-breaker.json\"\n  rm -f \"$OMC_STATE\u002Fsessions\u002F$SESSION_ID\u002Fskill-active-state.json\"\n  # Write cancel signal so stop hook detects cancellation in progress\n  NOW_ISO=\"$(date -u +\"%Y-%m-%dT%H:%M:%SZ\")\"\n  EXPIRES_ISO=\"$(date -u -d \"+30 seconds\" +\"%Y-%m-%dT%H:%M:%SZ\" 2>\u002Fdev\u002Fnull || python3 - \u003C\u003C'PY'\\nfrom datetime import datetime, timedelta, timezone\\nprint((datetime.now(timezone.utc) + timedelta(seconds=30)).strftime('%Y-%m-%dT%H:%M:%SZ'))\\nPY\\n)\"\n  printf '{\"active\":true,\"requested_at\":\"%s\",\"expires_at\":\"%s\",\"mode\":\"%s\",\"source\":\"bash_fallback\"}' \\\n    \"$NOW_ISO\" \"$EXPIRES_ISO\" \"$MODE\" > \"$OMC_STATE\u002Fsessions\u002F$SESSION_ID\u002Fcancel-signal-state.json\"\nfi\n\n# Clear legacy state only if no session ID (avoid clearing another session's state)\nif [ -z \"$SESSION_ID\" ]; then\n  rm -f \"$OMC_STATE\u002F${MODE}-state.json\"\nfi\n```\n\n## Auto-Detection\n\n`\u002Foh-my-claudecode:cancel` follows the session-aware state contract:\n- By default the command inspects the current session via `state_list_active` and `state_get_status`, navigating `.omc\u002Fstate\u002Fsessions\u002F{sessionId}\u002F…` to discover which mode is active.\n- When a session id is provided or already known, that session-scoped path is authoritative. Legacy files in `.omc\u002Fstate\u002F*.json` are consulted only as a compatibility fallback if the session id is missing or empty.\n- Swarm is a shared SQLite\u002Fmarker mode (`.omc\u002Fstate\u002Fswarm.db` \u002F `.omc\u002Fstate\u002Fswarm-active.marker`) and is not session-scoped.\n- The default cleanup flow calls `state_clear` with the session id to remove only the matching session files; modes stay bound to their originating session.\n\nActive modes are still cancelled in dependency order:\n1. Autopilot (includes linked ralph\u002Fultraqa\u002F cleanup)\n2. Ralph (cleans its linked ultrawork or )\n3. Ultrawork (standalone)\n4. UltraQA (standalone)\n5. Swarm (standalone)\n6. Ultrapilot (standalone)\n7. Pipeline (standalone)\n8. Team (Claude Code native)\n9. OMC Teams (tmux CLI workers)\n10. Plan Consensus (standalone)\n11. Self-Improve (standalone — clear state, clean orphaned worktrees, preserve iteration_state for resume, set status: \"user_stopped\" in the resolved `\u003Cself-improve-root>\u002Fstate\u002Fagent-settings.json`; new runs use `.omc\u002Fself-improve\u002Ftopics\u002F\u003Ctopic-slug>\u002F`, with flat `.omc\u002Fself-improve\u002F` retained only for legacy single-track resumes)\n\n## Force Clear All\n\nUse `--force` or `--all` when you need to erase every session plus legacy artifacts, e.g., to reset the workspace entirely.\n\n```\n\u002Foh-my-claudecode:cancel --force\n```\n\n```\n\u002Foh-my-claudecode:cancel --all\n```\n\nSteps under the hood:\n1. `state_list_active` enumerates `.omc\u002Fstate\u002Fsessions\u002F{sessionId}\u002F…` to find every known session.\n2. `state_clear` runs once per session to drop that session’s files.\n3. A global `state_clear` without `session_id` removes legacy files under `.omc\u002Fstate\u002F*.json`, `.omc\u002Fstate\u002Fswarm*.db`, and compatibility artifacts (see list).\n4. Team artifacts (`~\u002F.claude\u002Fteams\u002F*\u002F`, `~\u002F.claude\u002Ftasks\u002F*\u002F`, `.omc\u002Fstate\u002Fteam-state.json`) are best-effort cleared as part of the legacy fallback.\n   - Cancel for native team does NOT affect omc-teams state, and vice versa.\n\nEvery `state_clear` command honors the `session_id` argument, so even force mode still uses the session-aware paths first before deleting legacy files.\n\nLegacy compatibility list (removed only under `--force`\u002F`--all`):\n- `.omc\u002Fstate\u002Fautopilot-state.json`\n- `.omc\u002Fstate\u002Fralph-state.json`\n- `.omc\u002Fstate\u002Fralph-plan-state.json`\n- `.omc\u002Fstate\u002Fralph-verification.json`\n- `.omc\u002Fstate\u002Fultrawork-state.json`\n- `.omc\u002Fstate\u002Fultraqa-state.json`\n- `.omc\u002Fstate\u002Fswarm.db`\n- `.omc\u002Fstate\u002Fswarm.db-wal`\n- `.omc\u002Fstate\u002Fswarm.db-shm`\n- `.omc\u002Fstate\u002Fswarm-active.marker`\n- `.omc\u002Fstate\u002Fswarm-tasks.db`\n- `.omc\u002Fstate\u002Fultrapilot-state.json`\n- `.omc\u002Fstate\u002Fultrapilot-ownership.json`\n- `.omc\u002Fstate\u002Fpipeline-state.json`\n- `.omc\u002Fstate\u002Fomc-teams-state.json`\n- `.omc\u002Fstate\u002Fplan-consensus.json`\n- `.omc\u002Fstate\u002Fralplan-state.json`\n- `.omc\u002Fstate\u002Fboulder.json`\n- `.omc\u002Fstate\u002Fhud-state.json`\n- `.omc\u002Fstate\u002Fsubagent-tracking.json`\n- `.omc\u002Fstate\u002Fsubagent-tracker.lock`\n- `.omc\u002Fstate\u002Frate-limit-daemon.pid`\n- `.omc\u002Fstate\u002Frate-limit-daemon.log`\n- `.omc\u002Fstate\u002Fcheckpoints\u002F` (directory)\n- `.omc\u002Fstate\u002Fsessions\u002F` (empty directory cleanup after clearing sessions)\n\n## Implementation Steps\n\nWhen you invoke this skill:\n\n### 1. Parse Arguments\n\n```bash\n# Check for --force or --all flags\nFORCE_MODE=false\nif [[ \"$*\" == *\"--force\"* ]] || [[ \"$*\" == *\"--all\"* ]]; then\n  FORCE_MODE=true\nfi\n```\n\n### 2. Detect Active Modes\n\nThe skill now relies on the session-aware state contract rather than hard-coded file paths:\n1. Call `state_list_active` to enumerate `.omc\u002Fstate\u002Fsessions\u002F{sessionId}\u002F…` and discover every active session.\n2. For each session id, call `state_get_status` to learn which mode is running (`autopilot`, `ralph`, `ultrawork`, etc.) and whether dependent modes exist.\n3. If a `session_id` was supplied to `\u002Foh-my-claudecode:cancel`, skip legacy fallback entirely and operate solely within that session path; otherwise, consult legacy files in `.omc\u002Fstate\u002F*.json` only if the state tools report no active session. Swarm remains a shared SQLite\u002Fmarker mode outside session scoping.\n4. Any cancellation logic in this doc mirrors the dependency order discovered via state tools (autopilot → ralph → …).\n\n### 3A. Force Mode (if --force or --all)\n\nUse force mode to clear every session plus legacy artifacts via `state_clear`. Direct file removal is reserved for legacy cleanup when the state tools report no active sessions.\n\n### 3B. Smart Cancellation (default)\n\n#### If Team Active (Claude Code native)\n\nTeams are detected by checking for config files in `${CLAUDE_CONFIG_DIR:-~\u002F.claude}\u002Fteams\u002F`:\n\n```bash\n# Check for active teams\nTEAM_CONFIGS=$(find \"${CLAUDE_CONFIG_DIR:-$HOME\u002F.claude}\"\u002Fteams -name config.json -maxdepth 2 2>\u002Fdev\u002Fnull)\n```\n\n**Two-pass cancellation protocol:**\n\n**Pass 1: Graceful Shutdown**\n```\nFor each team found in ${CLAUDE_CONFIG_DIR:-~\u002F.claude}\u002Fteams\u002F:\n  1. Read config.json to get team_name and members list\n  2. For each non-lead member:\n     a. Send shutdown_request via SendMessage\n     b. Wait up to 15 seconds for shutdown_response\n     c. If response received: member terminates and is auto-removed\n     d. If timeout: mark member as unresponsive, continue to next\n  3. Log: \"Graceful pass: X\u002FY members responded\"\n```\n\n**Pass 2: Reconciliation**\n```\nAfter graceful pass:\n  1. Re-read config.json to check remaining members\n  2. If only lead remains (or config is empty): proceed to TeamDelete\n  3. If unresponsive members remain:\n     a. Wait 5 more seconds (they may still be processing)\n     b. Re-read config.json again\n     c. If still stuck: attempt TeamDelete anyway\n     d. If TeamDelete fails: report manual cleanup path\n```\n\n**TeamDelete + Cleanup:**\n```\n  1. Call TeamDelete() — removes ~\u002F.claude\u002Fteams\u002F{name}\u002F and ~\u002F.claude\u002Ftasks\u002F{name}\u002F\n  2. Clear team state: state_clear(mode=\"team\")\n  3. Check for linked ralph: state_read(mode=\"ralph\") — if linked_team is true:\n     a. Clear ralph state: state_clear(mode=\"ralph\")\n     b. Clear linked ultrawork if present: state_clear(mode=\"ultrawork\")\n  4. Run orphan scan (see below)\n  5. Emit structured cancel report\n```\n\n**Orphan Detection (Post-Cleanup):**\n\nAfter TeamDelete, verify no agent processes remain:\n```bash\nnode \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fcleanup-orphans.mjs\" --team-name \"{team_name}\"\n```\n\nThe orphan scanner:\n1. Checks `ps aux` (Unix) or `tasklist` (Windows) for processes with `--team-name` matching the deleted team\n2. For each orphan whose team config no longer exists: sends SIGTERM, waits 5s, sends SIGKILL if still alive\n3. Reports cleanup results as JSON\n\nUse `--dry-run` to inspect without killing. The scanner is safe to run multiple times.\n\n**Structured Cancel Report:**\n```\nTeam \"{team_name}\" cancelled:\n  - Members signaled: N\n  - Responses received: M\n  - Unresponsive: K (list names if any)\n  - TeamDelete: success\u002Ffailed\n  - Manual cleanup needed: yes\u002Fno\n    Path: ~\u002F.claude\u002Fteams\u002F{name}\u002F and ~\u002F.claude\u002Ftasks\u002F{name}\u002F\n```\n\n**Implementation note:** The cancel skill is executed by the LLM, not as a bash script. When you detect an active team:\n1. Read `${CLAUDE_CONFIG_DIR:-~\u002F.claude}\u002Fteams\u002F*\u002Fconfig.json` to find active teams\n2. If multiple teams exist, cancel oldest first (by `createdAt`)\n3. For each non-lead member, call `SendMessage(type: \"shutdown_request\", recipient: member-name, content: \"Cancelling\")`\n4. Wait briefly for shutdown responses (15s per member timeout)\n5. Re-read config.json to check for remaining members (reconciliation pass)\n6. Call `TeamDelete()` to clean up\n7. Clear team state: `state_clear(mode=\"team\", session_id)`\n8. Report structured summary to user\n\n#### If Autopilot Active\n\nAutopilot handles its own cleanup including linked ralph and ultraqa.\n\n1. Read autopilot state via `state_read(mode=\"autopilot\", session_id)` to get current phase\n2. Check for linked ralph via `state_read(mode=\"ralph\", session_id)`:\n   - If ralph is active and has `linked_ultrawork: true`, clear ultrawork first: `state_clear(mode=\"ultrawork\", session_id)`\n   - Clear ralph: `state_clear(mode=\"ralph\", session_id)`\n3. Check for linked ultraqa via `state_read(mode=\"ultraqa\", session_id)`:\n   - If active, clear it: `state_clear(mode=\"ultraqa\", session_id)`\n4. Mark autopilot inactive (preserve state for resume) via `state_write(mode=\"autopilot\", session_id, state={active: false, ...existing})`\n\n#### If Ralph Active (but not Autopilot)\n\n1. Read ralph state via `state_read(mode=\"ralph\", session_id)` to check for linked ultrawork\n2. If `linked_ultrawork: true`:\n   - Read ultrawork state to verify `linked_to_ralph: true`\n   - If linked, clear ultrawork: `state_clear(mode=\"ultrawork\", session_id)`\n3. Clear ralph: `state_clear(mode=\"ralph\", session_id)`\n\n#### If Ultrawork Active (standalone, not linked)\n\n1. Read ultrawork state via `state_read(mode=\"ultrawork\", session_id)`\n2. If `linked_to_ralph: true`, warn user to cancel ralph instead (which cascades)\n3. Otherwise clear: `state_clear(mode=\"ultrawork\", session_id)`\n\n#### If UltraQA Active (standalone)\n\nClear directly: `state_clear(mode=\"ultraqa\", session_id)`\n\n#### No Active Modes\n\nReport: \"No active OMC modes detected. Use --force to clear all state files anyway.\"\n\n## Implementation Notes\n\nThe cancel skill runs as follows:\n1. Parse the `--force` \u002F `--all` flags, tracking whether cleanup should span every session or stay scoped to the current session id.\n2. Use `state_list_active` to enumerate known session ids and `state_get_status` to learn the active mode (`autopilot`, `ralph`, `ultrawork`, etc.) for each session.\n3. When operating in default mode, call `state_clear` with that session_id to remove only the session’s files, then run mode-specific cleanup (autopilot → ralph → …) based on the state tool signals.\n4. In force mode, iterate every active session, call `state_clear` per session, then run a global `state_clear` without `session_id` to drop legacy files (`.omc\u002Fstate\u002F*.json`, compatibility artifacts) and report success. Swarm remains a shared SQLite\u002Fmarker mode outside session scoping.\n5. Team artifacts (`~\u002F.claude\u002Fteams\u002F*\u002F`, `~\u002F.claude\u002Ftasks\u002F*\u002F`, `.omc\u002Fstate\u002Fteam-state.json`) remain best-effort cleanup items invoked during the legacy\u002Fglobal pass.\n6. **Always** clear skill-active state as the final step, regardless of which mode was active or whether `--force` was used:\n   ```\n   state_clear(mode=\"skill-active\", session_id)\n   ```\n   This ensures the stop hook does not keep firing skill-protection reinforcements after cancel due to a stale `skill-active-state.json`. See issue #2118.\n\nState tools always honor the `session_id` argument, so even force mode still clears the session-scoped paths before deleting compatibility-only legacy state.\n\nMode-specific subsections below describe what extra cleanup each handler performs after the state-wide operations finish.\n## Messages Reference\n\n| Mode | Success Message |\n|------|-----------------|\n| Autopilot | \"Autopilot cancelled at phase: {phase}. Progress preserved for resume.\" |\n| Ralph | \"Ralph cancelled. Persistent mode deactivated.\" |\n| Ultrawork | \"Ultrawork cancelled. Parallel execution mode deactivated.\" |\n| UltraQA | \"UltraQA cancelled. QA cycling workflow stopped.\" |\n| Swarm | \"Swarm cancelled. Coordinated agents stopped.\" |\n| Ultrapilot | \"Ultrapilot cancelled. Parallel autopilot workers stopped.\" |\n| Pipeline | \"Pipeline cancelled. Sequential agent chain stopped.\" |\n| Team | \"Team cancelled. Teammates shut down and cleaned up.\" |\n| Plan Consensus | \"Plan Consensus cancelled. Planning session ended.\" |\n| Force | \"All OMC modes cleared. You are free to start fresh.\" |\n| None | \"No active OMC modes detected.\" |\n\n## What Gets Preserved\n\n| Mode | State Preserved | Resume Command |\n|------|-----------------|----------------|\n| Autopilot | Yes (phase, files, spec, plan, verdicts) | `\u002Foh-my-claudecode:autopilot` |\n| Ralph | No | N\u002FA |\n| Ultrawork | No | N\u002FA |\n| UltraQA | No | N\u002FA |\n| Swarm | No | N\u002FA |\n| Ultrapilot | No | N\u002FA |\n| Pipeline | No | N\u002FA |\n| Plan Consensus | Yes (plan file path preserved) | N\u002FA |\n\n## Notes\n\n- **Dependency-aware**: Autopilot cancellation cleans up Ralph and UltraQA\n- **Link-aware**: Ralph cancellation cleans up linked Ultrawork\n- **Safe**: Only clears linked Ultrawork, preserves standalone Ultrawork\n- **Local-only**: Clears state files in `.omc\u002Fstate\u002F` directory\n- **Resume-friendly**: Autopilot state is preserved for seamless resume\n- **Team-aware**: Detects native Claude Code teams and performs graceful shutdown\n\n## MCP Worker Cleanup\n\nWhen cancelling modes that may have spawned MCP workers (team bridge daemons), the cancel skill should also:\n\n1. **Check for active MCP workers**: Look for heartbeat files at `.omc\u002Fstate\u002Fteam-bridge\u002F{team}\u002F*.heartbeat.json`\n2. **Send shutdown signals**: Write shutdown signal files for each active worker\n3. **Kill tmux sessions**: Run `tmux kill-session -t omc-team-{team}-{worker}` for each worker\n4. **Clean up heartbeat files**: Remove all heartbeat files for the team\n5. **Clean up shadow registry**: Remove `.omc\u002Fstate\u002Fteam-mcp-workers.json`\n\n### Force Clear Addition\n\nWhen `--force` is used, also clean up:\n```bash\nrm -rf .omc\u002Fstate\u002Fteam-bridge\u002F       # Heartbeat files\nrm -f .omc\u002Fstate\u002Fteam-mcp-workers.json  # Shadow registry\n# Kill all omc-team-* tmux sessions\ntmux list-sessions -F '#{session_name}' 2>\u002Fdev\u002Fnull | grep '^omc-team-' | while read s; do tmux kill-session -t \"$s\" 2>\u002Fdev\u002Fnull; done\n```\n",{"name":5,"level":23,"description":7,"argument-hint":24},2,"[--force|--all]","main","f37de8f681f0c0ca0e0ebab6e5fddf764ca7553d","skills\u002Fcancel",1,"https:\u002F\u002Fgithub.com\u002Fagisota\u002Fold-one","https:\u002F\u002Fagisota.github.io\u002Fold-one\u002F","https:\u002F\u002Fgithub.com\u002Fagisota","pzd.world","2026-07-04 22:32:06",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add agisota\u002Fold-one --skill cancel",{"tool":40,"command":41},"bunx","bunx skills add agisota\u002Fold-one --skill cancel",{"tool":43,"command":44},"pnpm","pnpm dlx skills add agisota\u002Fold-one --skill cancel","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fagisota\u002Fold-one\u002Fblob\u002Fmain\u002Fskills\u002Fcancel\u002FSKILL.md",true,[],[49,56,63,72,79,86],{"skillNo":50,"name":51,"nameZh":6,"description":52,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":53,"categories":54,"keywords":55,"publishedAt":18},"01M75KRGZWQWE9JDBS6RX42NWV","zoom-out","Tell the agent to zoom out and give broader context or a higher-level perspective. Use when you're unfamiliar with a section of code or need to understand how it fits into the bigger picture.","https:\u002F\u002Fgithub.com\u002Fagisota\u002Fold-one\u002Fblob\u002Fmain\u002Fskills\u002Fzoom-out\u002FSKILL.md",[],[],{"skillNo":57,"name":58,"nameZh":6,"description":59,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":18},"01M75KRCZHKE0ZYAQFFYHS0YNP","zeroize-audit","Detects missing zeroization of sensitive data in source code and identifies zeroization removed by compiler optimizations, with assembly-level analysis, and control-flow verification. Use for auditing C\u002FC++\u002FRust code handling secrets, keys, passwords, or other sensitive data.","https:\u002F\u002Fgithub.com\u002Fagisota\u002Fold-one\u002Fblob\u002Fmain\u002Fskills\u002Fzeroize-audit\u002FSKILL.md",[],[],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":68,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":18},"01M75KR94KYQ30DTE1P9WNDAE0","yeet","Use only when the user explicitly asks to stage, commit, push, and open a GitHub pull request in one flow using the GitHub CLI (`gh`).","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fagisota\u002Fold-one\u002Fblob\u002Fmain\u002Fskills\u002Fyeet\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fagisota\u002Fold-one\u002Fblob\u002Fmain\u002Fskills\u002Fyeet\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":18},"01M75KR3H13YJ9QHMD49818FA1","yara-rule-authoring","Guides authoring of high-quality YARA-X detection rules for malware identification. Use when writing, reviewing, or optimizing YARA rules. Covers naming conventions, string selection, performance optimization, migration from legacy YARA, and false positive reduction. Triggers on: YARA, YARA-X, malware detection, threat hunting, IOC, signature, crx module, dex module.","https:\u002F\u002Fgithub.com\u002Fagisota\u002Fold-one\u002Fblob\u002Fmain\u002Fskills\u002Fyara-rule-authoring\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":18},"01M75KQMPY8A4Y8XTDD2DT0M9T","x-twitter-growth","X\u002FTwitter growth engine for building audience, crafting viral content, and analyzing engagement. Use when the user wants to grow on X\u002FTwitter, write tweets or threads, analyze their X profile, research competitors on X, plan a posting strategy, or optimize engagement. Complements social-content (generic multi-platform) with X-specific depth: algorithm mechanics, thread engineering, reply strategy, profile optimization, and competitive intelligence via web search.","https:\u002F\u002Fgithub.com\u002Fagisota\u002Fold-one\u002Fblob\u002Fmain\u002Fskills\u002Fx-twitter-growth\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":18},"01M75KQHNZ0ANZNJGZY6M8MDAK","x-api","X\u002FTwitter API integration for posting tweets, threads, reading timelines, search, and analytics. Covers OAuth auth patterns, rate limits, and platform-native content posting. Use when the user wants to interact with X programmatically.","https:\u002F\u002Fgithub.com\u002Fagisota\u002Fold-one\u002Fblob\u002Fmain\u002Fskills\u002Fx-api\u002FSKILL.md",[],[]]