retro
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests untrusted data from coding session logs and repository metadata, which then informs modifications to steering files like
CODING_STANDARDS.mdorAGENTS.md. - Ingestion points: Coding session logs and primary session sources (SKILL.md, Step 2).
- Boundary markers: The instructions do not define specific delimiters or warnings for the agent when processing external log data.
- Capability inventory: The skill reads files, searches logs, and modifies steering files or repository documentation (SKILL.md, Step 3 and References section).
- Sanitization: No sanitization or validation of the ingested log content is performed before analysis or file modification.
Audit Metadata