OpenParable
Skip to content
← Back to skills

Tracking Doc Worker

Productivity
tracking-doc-worker

Tracking Doc Worker is an Agent Skill for productivity published in the CorrectRoadH/OpenTickly repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Audit and finalize inline source-document status blocks for the timer refactor mission.

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/CorrectRoadH/OpenTickly/blob/HEAD/.factory/skills/tracking-doc-worker/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/tracking-doc-worker, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

Tracking Doc Worker

NOTE: Startup and cleanup are handled by worker-base. This skill defines the work procedure.

When to Use This Skill

Use for features that primarily update or audit the inline status blocks in the closed set of implementation-driving source docs.

Required Skills

None.

Work Procedure

  1. Read mission mission.md, mission AGENTS.md, .factory/library/documentation-traceability.md, and the current features.json.
  2. Treat the closed source-doc list in mission AGENTS.md as canonical. Do not audit docs outside that list unless the orchestrator changes it first.
  3. Read the relevant completed feature handoffs or verification evidence so the status blocks describe the real final implementation state.
  4. Add or update exactly one active mission status block in each listed source doc.
  5. Use the canonical field labels exactly:
    • Agent
    • Status
    • Current code differences
    • Todo
    • Fully implemented
  6. Keep the blocks honest:
    • if work remains, do not mark Fully implemented: yes
    • if Todo is none, the remaining fields must also describe a landed state
  7. Run the traceability validation command from .factory/services.yaml after the edits.
  8. In the handoff, explicitly state:
    • which docs were updated
    • whether any listed doc still has open gaps
    • why each Fully implemented value is truthful

Example Handoff

{
  "salientSummary": "Updated the inline mission status blocks across the closed timer source-doc set and verified the canonical fields and consistency rules. The blocks now reflect which timer docs are fully landed versus which still describe open implementation gaps.",
  "whatWasImplemented": "Added or refreshed one active mission status block in each directly used timer source document listed in mission `AGENTS.md`, using the canonical field schema and aligning every block with the verified implementation state from the completed features. Ran the repo traceability check to confirm field presence and block coverage.",
  "whatWasLeftUndone": "",
  "verification": {
    "commandsRun": [
      {
        "command": "rg -n \"Mission status|Agent:|Status:|Current code differences:|Todo:|Fully implemented:\" docs/product/tracking.md docs/research/toggl-timer-*.md",
        "exitCode": 0,
        "observation": "Every listed source doc exposed one mission status block with the canonical field names."
      }
    ],
    "interactiveChecks": [
      {
        "action": "Manually reviewed each listed source doc against the final feature evidence.",
        "observed": "The status blocks matched the verified implementation state and no doc marked itself fully implemented while still listing open todo items."
      }
    ]
  },
  "tests": {
    "added": []
  },
  "discoveredIssues": []
}

When to Return to Orchestrator

  • A required source doc is missing from the closed list in mission AGENTS.md
  • The final implementation evidence is too incomplete to write truthful status blocks
  • The task reveals contradictions between completed feature claims and the actual landed implementation state