drill
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting and processing untrusted external data.
- Ingestion points: The skill reads project context including code, documentation, and recent commit messages (SKILL.md).
- Boundary markers: There are no explicit instructions or delimiters defined to isolate or treat ingested content as untrusted data.
- Capability inventory: The skill writes to local files including CONTEXT.md, ADRs, and implementation specifications (docs/specs/*.md). It also invokes high-capability skills like implement and prototype based on the content of these generated specifications.
- Sanitization: No sanitization or validation mechanisms are described for the data ingested from the project environment before it is interpolated into the design session or final specifications.
Audit Metadata