Menu

Individual skill

skill-vetting

0upvotes

Vet an agent skill before installing it — read the SKILL.md and any scripts for the red-flag patterns (credential access, obfuscation, exfiltration, prompt injection), audit its blast radius, and produce a risk-tiered verdict. Use when asked is this skill safe to install, vet this SKILL.md, review this skill from a marketplace, or check what this skill can do to my machine. Produces the risk classification with quoted evidence, the permission-surface audit, the red-flag checklist results, and an install/sandbox/reject recommendation.

plugins/pm-security/skills/skill-vetting/SKILL.md

Read pinned source

Source commit f7c90bc09ee05dc25cf257bdb1826f2c2552c2cb

What was checked

Matches the observed repository HEAD
Yes
Scanner rules are current
Yes
Repository scan complete
No
Skill scan complete
No
Observed behavior
Network or broad shell
Last checked
Nov 9, 2026

These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.

Install this skill

Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.

An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.

Repository context

Repository stars (not skill ratings)
1,538
Repository last updated
Nov 9, 2026
License
MIT

Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.

Files and excerpt

File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.

Browse files
Read excerpt

# Skill Vetting Skill

A skill is instructions your agent will *obey* plus scripts your machine will *run* — installing one is granting authorship over future behavior, and marketplaces host both gems and traps. This skill is the pre-install reading: the red-flag pattern sweep, the blast-radius audit ("what can this touch"), and a tiered verdict with quoted evidence. It's a judgment framework, not a scanner — the point is an informed human decision, and for anything above LOW the human makes it.

## What This Skill Produces

- **The verdict** — 🟢 LOW / 🟡 MEDIUM / 🔴 HIGH / ⛔ REJECT, with the one-paragraph reasoning
- **The evidence table** — every finding with the quoted line from the skill's own files
- **The blast-radius audit** — files read/written, network destinations, commands run, credentials touched
- **The recommendation** — install / install-but-watch / sandbox first / reject, matched to the tier

## Required Inputs

Ask for these if not provided:
Back to top