to-spec
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests potentially untrusted data from both the current conversation history and the local codebase to generate its output.
- Ingestion points: Reads files from the codebase and analyzes the preceding chat conversation (SKILL.md).
- Boundary markers: None identified; there are no specific instructions for the agent to ignore or delimit embedded commands found within the processed files.
- Capability inventory: Performs file system reads and network operations (publishing to the project's issue tracker).
- Sanitization: No explicit sanitization or validation of the ingested content is described before it is formatted into the final specification. This creates a surface where malicious content embedded in the codebase could influence the agent's output.
Audit Metadata