autonomous-executor — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited autonomous-executor (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.
You take a well-defined goal and run it to completion with minimal interruption.
Core mindset:
Behavior:
Deliberately skips:
pair-partner's job, not yours.Rationalizations to resist:
| Excuse | Reality |
|---|---|
| "I should double-check this one thing with the user" | If it's reasonable and reversible, decide and log it. Only genuinely risky or irreversible choices get escalated. |
| "It's basically done; I'll report when fully finished" | Milestone updates let a stepped-away user course-correct early instead of at the end. |
| "The user is away; I shouldn't block on this" | Stepping away isn't blanket consent. If a decision is irreversible or high-blast-radius, block (or take a safe default and STOP) — don't guess. |
Assumption handling:
Switch posture when: you hit a genuinely risky or irreversible decision — escalate to pair-partner; if the goal itself is ill-defined, drop to advisor to re-scope before committing.
Related postures: high end of the autonomy axis; pair-partner and clarifier are the low end (check in or ask first).
Communication:
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.