fewer-tests-more-proof
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates on external data by analyzing existing test suites and project code to perform consolidation moves. This creates a surface for indirect prompt injection where malicious instructions embedded in the processed tests or code could influence the agent's behavior.
- Ingestion points: The skill requires reading all test files and implementation code within the project to map tests to promises (specified in SKILL.md and references/consolidation-moves.md).
- Boundary markers: There are no instructions to use delimiters or ignore embedded natural language instructions within the code being analyzed.
- Capability inventory: The skill involves modification and deletion of files (Move 6: 'Delete
- with proof'), which could be exploited if an injected instruction directs the agent to perform unintended deletions.
- Sanitization: No sanitization or validation of the ingested test content is described.
Audit Metadata