qe-release
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to grep the entire codebase for version patterns and process the results. This creates a surface where untrusted content from files in the repository could be ingested into the agent's context as instructions.
- Ingestion points: Grep output from the entire codebase (Step 5).
- Boundary markers: Absent; no instructions provided to ignore or delimit embedded text in search results.
- Capability inventory: Writing to package configuration files, updating changelogs, and invoking quality assessment tools via the @agentic-qe MCP server.
- Sanitization: Absent; no validation or escaping is performed on the data retrieved from the codebase.
Audit Metadata