mir-review-snippet
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill executes
npx mir sync <name>andnpx mir install <name>. If the snippet name variable (<name>), which is derived from local files or metadata, contains shell metacharacters (e.g.,;,&,|), it could lead to arbitrary command execution. - [EXTERNAL_DOWNLOADS]: The use of
npx mirinvolves thenpxpackage runner, which may download themirpackage from the npm registry if it is not present in the local environment. - [PROMPT_INJECTION]: The skill reads and analyzes files within
.mir/snippets/. Since these files can contain arbitrary text, they serve as a vector for indirect prompt injection, where malicious instructions inside a snippet could attempt to subvert the agent's review logic. - Ingestion points:
.mir/snippets/<name>.yamland directory contents (SKILL.md) - Boundary markers: Absent
- Capability inventory: Local file reading, file modification, and shell command execution (SKILL.md)
- Sanitization: No explicit sanitization or validation of snippet content is mentioned before processing or inclusion in commands.
Audit Metadata