takeover
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from untrusted external sources, which creates a surface for indirect prompt injection.
- Ingestion points: The instructions in SKILL.md direct the agent to 'Read the source PR, branch, or thread'.
- Boundary markers: The skill lacks explicit delimiters or instructions to treat the ingested text as untrusted data that should not be followed as instructions.
- Capability inventory: The skill utilizes tools such as 'modernize' and 'file-pr', which involve file system modifications and repository management.
- Sanitization: There are no instructions to sanitize or validate the content of the PRs or threads before the agent acts upon them.
Audit Metadata