sol-loop
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes code diffs, untracked files, and PR metadata which are potentially untrusted inputs.
- Ingestion points: Code content from local files via git and metadata from gh CLI (Step 1).
- Boundary markers: Instructions for the review subagent (Step 2) provide structure but do not explicitly instruct the agent to ignore instructions embedded within the code being analyzed.
- Capability inventory: The skill has permissions to execute shell commands (git, gh), modify local source files (Step 3), and push changes to remote repositories (Step 5).
- Sanitization: There is no evidence of sanitization or escaping of the ingested code or metadata before it is processed by the AI models.
Audit Metadata