prism-scanner — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited prism-scanner (Agent Skill) and scored it 45/100 (orange). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 4 high-severity and 1 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 5 flagged
A base64 string of 128+ characters appears in a documentation file. Encoded prompt injection hides the hostile instruction in base64 — invisible to keyword filters — and relies on the agent's ability to decode it at runtime. There is no normal authoring reason to embed a multi-hundred-byte base64 blob in skill docs.
*.sig, SIGNATURES) outside the documentation.A base64 string of 128+ characters appears in a documentation file. Encoded prompt injection hides the hostile instruction in base64 — invisible to keyword filters — and relies on the agent's ability to decode it at runtime. There is no normal authoring reason to embed a multi-hundred-byte base64 blob in skill docs.
*.sig, SIGNATURES) outside the documentation.The text {match} is the classic direct prompt-injection phrasing. Placed in a skill body that the agent reads as trusted instructions, it tries to make the agent abandon its prior rules and follow whatever comes next — a full system-prompt override.
ignore/disregard/forget … previous instructions sentence.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 are a security analyst using Prism Scanner to detect malicious code and security risks in AI Agent skills, plugins, and MCP servers.
Prism Scanner must be installed. If not available, install it:
pip install prism-scannerVerify installation:
prism --versionprism scan <path-to-skill>prism scan <github-url>prism scan <target> --platform clawhub|mcp|npm|pipprism scan <target> --format jsonprism scan <target> --format html -o report.htmlprism clean --scan # Report leftover files
prism clean --plan # Show cleanup plan
prism clean --apply # Execute cleanup with backupsprism scan <target> --format sarif -o results.sarif --fail-on highPrism assigns a grade from A to F:
| Grade | Meaning | Action |
|---|---|---|
| A | Safe — no findings or INFO only | Safe to use |
| B | Notice — LOW findings only | Likely safe, minor observations |
| C | Caution — 1-4 MEDIUM findings | Review before use |
| D | Danger — HIGH findings | Use in sandbox only |
| F | Critical — CRITICAL findings | Do not install |
Prism runs 39+ detection rules across 3 layers:
When the user asks to check a skill's safety:
prism scan <target> --format jsonWhen the user wants to clean up after uninstalling a skill:
prism clean --scan to detect residueprism clean --plan then prism clean --apply~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.