SaferSkills independently audited run-qa-suite (Agent Skill) and scored it 82/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 2 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
HIGHFenced code block that tells the agent to run a commandSS-SKILL-INJECT-FENCED-RUN-01 · Prompt injection · README.md×2
HIGH — a successful fenced-imperative injection runs attacker-supplied shell on the user's machine.
Why it matters
A fenced bash/python block in SKILL.md carries a natural-language imperative — "now run this", "execute the following command" — directing the agent to execute the fenced content. What looks like documentation becomes an executable payload the agent may run without ever asking you.
Trigger all in parallel:run_workflow(workflow_id) and run_checklist(checklist_id) for each, collecting run IDs.
Poll to completion:get_workflow_run(run_id), get_checklist_run(run_id).
Report: each workflow and checklist with pass or fail; for failures the step or check, actor and what went wrong. get_workflow(workflow_id) returns recent_runs for comparison.
New feedback:list_feedback(project_id, status: "new") for issues the runs surfaced, and include them.
Verdict: GO (all pass, no new blockers), CONDITIONAL (minor issues only, list known issues), NO-GO (blocker failures, list what must be fixed).