retro
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from untrusted sources, including repository files and previous session logs, which creates a potential surface for indirect instructions.
- Ingestion points: The skill reads session logs from
~/.pi/agent/sessionsand primary source files from the current repository as described inSKILL.md. - Boundary markers: No specific boundary markers or delimiters are instructed to be used when interpolating these files into the context window.
- Capability inventory: The skill performs search and read operations to identify navigation issues, missing automated checks, and coding standard violations. It does not perform write operations or network requests within the skill logic.
- Sanitization: The instructions do not define sanitization or filtering procedures for the analyzed content.
Audit Metadata