openplan-record — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited openplan-record (Agent Skill) and scored it 96/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 1 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Use this after $openplan-core when conversation, review, or implementation output needs to become durable memory or documentation.
Memory and docs represent the human goal/intent. Do not record agent guesses as the human goal/intent. If the source is not explicit, mark it as inference or pending confirmation.
| Type | Purpose | Lifetime | Examples |
|---|---|---|---|
| Memory | Standing human preference or operating constraint | Long-lived, compact | "Do not checkout without permission", "docs in Chinese" |
| Docs | Project source of truth and rationale | Long-lived, structured | architecture, workflow, ADR, conventions |
| Handoff | Cross-agent/session continuity | Short/medium-lived | current objective, blockers, next safe action |
| Review | Evidence-backed quality finding | Snapshot | blocking issue, residual risk |
| HTML brief | Temporary human frontend | Disposable | visual summary, teaching page |
.openplan/, docs/, repo root, separate docs repo, or user-provided path;Human decisions required.When asked to update records, produce:
Record classification:
Proposed destination:
Destination root:
Source evidence:
Exact update:
Requires human confirmation: yes/no
Staleness/conflict notes:If editing is authorized, make the narrowest update and report the changed path.
Use $openplan-garden before broad documentation cleanup, and use it after cleanup when zero-context readability matters.
references/record-policy.md: detailed memory/docs decision policy.references/update-contract.md: update proposal and changelog contract.assets/claude-code-command.md: Claude Code prompt seed.~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.