SECURITY REVIEW / Sentry
Skill scanner
Audits a skill before install, with a bundled script and explicit confidence levels.
“Scan agent skills for security issues. Use when asked to "scan a skill", "audit a skill", "review skill security", "check skill for injection", "validate SKILL.md", or assess whether an agent skill is safe to install. Checks for prompt injection, malicious scripts, excessive permissions, secret exposure, and supply chain risks.”
01 / THE REASONING
Why this made the selection.
- A Bundled Script heading means the check is not only prose.
- Confidence Levels and an output format are separate headings.
02 / THE REVIEW RECORD
What we actually inspected.
Source review has boundaries.
A clear record is more useful than a “safe” badge.
Material inspected
- skills/skill-scanner/SKILL.md
- README.md
- LICENSE
- skills/skill-scanner/scripts/scan_skill.py
Our findings
- Headings include Bundled Script, Workflow and Confidence Levels.
- README documents npx skills add getsentry/skills.
- The scan_skill function was inspected: top-level definition, immediate references/scripts and separate structural checks feed the result. Unreadable referenced files can be skipped.
- Frontmatter checks require name and description, compare name with the directory, and question unrestricted tools.
- The behavioral-analysis section explicitly checks changes to agent instructions, memory, settings, MCP configuration, hooks, and auto-approval lists because those changes can persist after a skill is removed.
Not established by this review
- The skill was not executed inside a host agent.
- Not executed: it was not installed or run.
The review applies to the material and revision named here. A newer upstream release can change its behavior.
03 / PUT IT TO WORK
Add a skill to your workflow.
DOCUMENTED COMMAND
npx skills add getsentry/skills/skills/skill-scanner --skill skill-scannerRun in your terminal to choose your AI tools, project or global scope, and installation method. This short command follows the current upstream version; the review covers the revision shown above.
Optional: install the reviewed revision
Keeps the original source and installer version where recorded, while leaving your AI tools and installation scope for you to choose.
npx skills@1.7.1 add https://github.com/getsentry/skills/tree/d18b7aa8ba878354e5c348310230e652f7690f9c/skills/skill-scanner --skill skill-scanner- Choose your AI tools and project or global installation scope in Skills CLI. The review applies to the source revision shown on this page.
- Retain all supporting files; separately configure companion skills, runtimes and service credentials named in the definition.
- Review the resulting diff and use an isolated target before any migration, deployment, load test or warehouse operation.
Before you start
- uv and the full skill-scanner directory, including scripts and reference patterns.
04 / FOLLOW THE EVIDENCE
The source trail.
Our notes are separate from the original resource.
Check upstream before adopting a new version.
- Skill definition ↗Checked
https://github.com/getsentry/skills/blob/d18b7aa8ba878354e5c348310230e652f7690f9c/skills/skill-scanner/SKILL.md
Supports: Workflow and stated scope
- Repository licence ↗Checked
https://github.com/getsentry/skills/blob/d18b7aa8ba878354e5c348310230e652f7690f9c/LICENSE
Supports: Redistribution terms
- Install instructions ↗Checked
https://github.com/getsentry/skills/blob/d18b7aa8ba878354e5c348310230e652f7690f9c/README.md
Supports: Documented install path
https://github.com/getsentry/skills/blob/d18b7aa8ba878354e5c348310230e652f7690f9c/skills/skill-scanner/scripts/scan_skill.py
Supports: review, limitations, access, install
https://raw.githubusercontent.com/getsentry/skills/d18b7aa8ba878354e5c348310230e652f7690f9c/skills/skill-scanner/SKILL.md
Supports: review.findings, bestFor