Audit, optimize, and maintain your OpenClaw agent workspace. Use when workspace needs health check, skill audit, memory cleanup, or performance tuning.
Coding
OpenClaw Workspace Doctor
Try itUse when the user wants to inspect, repair, package, or troubleshoot an OpenClaw workspace doctor setup, especially for stale bootstrap state, missing daily...
What it does
Use when the user wants to inspect, repair, package, or troubleshoot an OpenClaw workspace doctor setup, especially for stale bootstrap state, missing daily...
The skill document
OpenClaw Workspace Doctor
Overview
This skill runs the OpenClaw workspace doctor, applies safe workspace-local fixes, and handles the companion fixer for the external codex-cli backend config. Use it when the user asks to diagnose workspace startup drift, package the doctor tool, or repair the known --full-auto non-interactive stall risk.
When To Use It
- The user asks to run or extend the workspace doctor tool.
- The workspace has stale bootstrap/onboarding behavior.
- Daily
memory/YYYY-MM-DD.mdnotes are missing or inconsistent. - OpenClaw is hanging because
codex-clilaunches with--full-auto. - The user wants the doctor packaged as a repo or a Codex skill.
Workflow
-
Run the doctor first:
python3 scripts/workspace_doctor.py -
If the problem is workspace-local, apply safe fixes:
python3 scripts/workspace_doctor.py --fix -
If the remaining problem is the external OpenClaw config, inspect or patch it with the companion fixer:
python3 scripts/fix_openclaw_codex_config.py --checkpython3 scripts/fix_openclaw_codex_config.py -
If the environment blocks writing
~/.openclaw/openclaw.json, do not pretend the live fix happened. Report the sandbox restriction clearly and, if useful, use:python3 scripts/fix_openclaw_codex_config.py --stdout -
For repo packaging tasks, update:
README.mdpyproject.tomlsetup.pytests/.github/workflows/ci.yml
-
Validate before closing:
python3 -m unittest discover -s testspython3 scripts/workspace_doctor.pypython3 scripts/workspace_doctor.py --jsonpython3 /Users/xutao/.codex/skills/.system/skill-creator/scripts/quick_validate.py skills/openclaw-workspace-doctor
Repo Commands
- Run doctor:
python3 scripts/workspace_doctor.py - Run doctor with fixes:
python3 scripts/workspace_doctor.py --fix - Emit JSON:
python3 scripts/workspace_doctor.py --json - Check external config:
python3 scripts/fix_openclaw_codex_config.py --check - Patch external config:
python3 scripts/fix_openclaw_codex_config.py - Print patched config without writing:
python3 scripts/fix_openclaw_codex_config.py --stdout
References
- Read references/commands.md for the command map and fix boundaries.
- Read references/repo-layout.md when you need the repo structure and what each area is for.
Scripts
scripts/run_doctor.pyruns the packaged CLI.scripts/fix_openclaw_config.pyruns the external config fixer.- Prefer these wrappers or the repo-root scripts instead of retyping logic.
Related skills
Set up an OpenClaw personal AI assistant by generating tailored configuration files through guided Q&A.
Navigate and use OpenClaw documentation efficiently with cached doc fetch/search tooling, sitemap/category routing, and config snippet lookup. Use when answe...
Create, analyze, and optimize OpenClaw workspace configuration files (SOUL.md, IDENTITY.md, USER.md, MEMORY.md, TOOLS.md, AGENTS.md, STYLE.md, HEARTBEAT.md) with guided workflows and best practices validation. Analyzes existing files for patterns, clarity, injection order, and content placement. All work is done in a sandbox area without modifying originals.
Route OpenClaw CLI requests to the right command family and profile, then verify the outcome.
Compress OpenClaw session context safely with backups and restore steps.