list-probes — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited list-probes (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
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.
A quick inventory of the probes in the attached session's memory. Use it to discover what you can pass to probe_eval, steer, or causality_protocol.
colab_attach, to see what is already loadedprobe_eval / steer / causality_protocol, to get a valid probe_id and its declared layerprobe_id — list to find the correct idCall the MCP tool `list_probes` (no arguments beyond session_name). Returns each probe's probe_id, model_id, layer, position, and source.
A short table, one row per probe:
{probe_id} — {model_id} · L{layer} · {position} · {source}Then suggest the natural next step (evaluate with probe_eval, or audit with causality_protocol).
list_probes. Passing the wrong layer to probe_evalsilently scores the wrong activations.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.