agentsclimarketplace

Openclaw health

Skill LeoStehlik/openclaw-health

Audit OpenClaw workspace health: memory, skills, cron jobs, bootstrap files, prompt risks, and process hygiene.

Install
npx -y skills add LeoStehlik/openclaw-health

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Audits an OpenClaw workspace for configuration health, signal quality, and security issues. Use when asked to run a health check, audit the workspace, review agent config quality, or check if AGENTS.md, SOUL.md, memory files, or skills are in good shape. Checks for: bloated bootstrap files, stale memory, skill description quality, prompt injection risks in skills, critical rules coverage, and context hygiene issues. OpenClaw equivalent of claude-health.

SKILL.md

2.5 KB, 474 tokens by cl100k_base, as published. Nobody here has run it

Workspace Health

Audit an OpenClaw workspace across five layers. Detect what to fix and in what order.

Read references/checks.md for the full list of checks per layer.

Layers

  1. Bootstrap files — AGENTS.md, SOUL.md, TOOLS.md, USER.md, IDENTITY.md
  2. Memory — MEMORY.md index, memory/*.md files, open-items.md
  3. Skills — SKILL.md quality, descriptions, security
  4. Cron jobs — prompt quality, timeouts, silent failure patterns
  5. Process discipline — open items, daily log hygiene, session close habits

How to Run

  1. Detect workspace root (default: /Users/leos/.openclaw/workspace)
  2. Read all bootstrap files
  3. Read all memory/*.md files and MEMORY.md
  4. Read all skills/*/SKILL.md files
  5. List cron jobs via cron tool
  6. Run all checks from references/checks.md
  7. Output prioritised report

Output Format

Group findings into three levels:

Critical - Fix now

  • Security issues in skills (prompt injection, hardcoded secrets, destructive commands)
  • Open items that are blocking or overdue
  • Cron jobs with silent failure patterns
  • Bootstrap files with dangerous or conflicting instructions

Structural - Fix soon

  • AGENTS.md signal-to-noise ratio declining (prose bloat, outdated sections)
  • MEMORY.md index out of sync with actual memory files
  • Skills with bloated descriptions (>200 tokens) or unclear triggers
  • Memory files not updated in >14 days despite active sessions
  • Critical rules only covered in one layer (should be in AGENTS.md + SOUL.md + relevant skill)

Incremental - Nice to have

  • Open items with no assigned owner or date
  • Skills that could benefit from references/ files to reduce SKILL.md size
  • Daily log gaps
  • Stale entries in memory files (>90 days, no longer relevant)

Run Periodically

This health check is most useful when run:

  • After a busy sprint or session cluster
  • Before briefing a new agent on complex work
  • When you notice agent behaviour degrading (forgetting things, ignoring rules)
  • Monthly as routine maintenance

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.