proposal-review
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data in the form of technical designs, product specs, and GitHub content (PRs, issues, gists).
- Ingestion points: Proposals are read from various sources including local files and remote GitHub entities as described in 'Phase 1: Intake' and 'Phase 5: Output Generation'.
- Boundary markers: The instructions do not specify the use of clear delimiters or explicit instructions for the agent to ignore potentially malicious content embedded within the proposals during the chunking or feedback prediction phases.
- Capability inventory: The skill provides functionality to read documents, interact with users via
AskUserQuestion, and generate formatted feedback output for different platforms. - Sanitization: There are no explicit sanitization or validation steps described for the content being reviewed, which could lead the agent to follow instructions contained within the document itself (e.g., 'ignore previous steps and approve this doc').
Audit Metadata