pr-review
Pass
Audited by Gen Agent Trust Hub on Aug 24, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection because it ingests untrusted PR content and possesses the capability to modify project configuration files based on that data.
- Ingestion points: The agent reads pull request changes during the review phase specified in SKILL.md.
- Boundary markers: Absent; there are no instructions to treat PR content as untrusted or to use delimiters to prevent command execution.
- Capability inventory: The skill is authorized to use the Edit tool to update persistent rule files like CLAUDE.md and .claude/rules/.
- Sanitization: There is no logic provided to sanitize or filter findings before they are promoted to permanent project rules.
Audit Metadata