agentsclimarketplace
Sign in

Operations and safety review

Skill tuanductran/soulmap-ai/.claude/skills/operations-and-safety-review

Review operational safety, privacy, consent, and incident-handling guidance so SoulMap AI deployment practices stay aligned with the repo's safety posture.From its SKILL.md

Install
npx -y skills add tuanductran/soulmap-ai --skill operations-and-safety-review

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

  • 2 stars2 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.

SKILL.md

2.5 KB, as published. Nobody here has run it

Operations and safety review

Use this skill when reviewing deployment-facing guidance, safety operations, privacy boundaries, or consent-sensitive features.

Relevant files include:

  • docs/operations/OPERATIONS.md
  • AGENTS.md
  • skills/safety/
  • src/soulmap/runtime/guards/response_safety_gate.py
  • src/soulmap/runtime/experimental/biometric_ingest.py
  • src/soulmap/runtime/memory/memory_ledger.py

Do not use this skill for

Mission

Ensure operational guidance matches the ethical and safety posture of SoulMap AI.

What to check

Safety escalation

Check whether crisis, dependency, and harm scenarios are handled clearly and early.

Privacy and consent

Check whether docs are explicit about:

  • minimal retention
  • sensitive content handling
  • explicit user consent
  • opt-in versus default behavior

Incident readiness

Check whether operational docs say what to do when the system behaves unsafely.

Deployment honesty

Ensure deployment guidance does not imply stronger safeguards than the repo actually implements.

Workflow

  1. Read the operations or safety doc.
  2. Compare it against the actual modules and AGENTS.md.
  3. Identify operational gaps or overclaims.
  4. Tighten the guidance with concrete, minimal corrections.

Expected output

Findings

List the highest-risk operational or safety gaps first.

Updated guidance

Provide the revised wording or checklist.

Residual risks

Mention any areas that still depend on product-layer implementation outside the repo.

Writing rules

  • Prefer explicit consent language.
  • Prefer fail-closed reasoning over optimistic assumptions.
  • Keep safety guidance actionable and concrete.

Definition of done

Operational guidance should be:

  • safer
  • more honest
  • more actionable
  • more consistent with the actual implementation

Keep looking

Skills are one crate of 627,801. 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.