writing-tests
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of instructional markdown files that guide an AI agent in software testing best practices. It does not contain any executable scripts, remote code download patterns, or hardcoded credentials.
- [INDIRECT_PROMPT_INJECTION]: While the skill inherently involves reading repository files to understand the testing context (runners, configurations, existing tests), it provides strong defensive guidelines for the agent. It emphasizes verifying actual behavior and explicitly states that it does not authorize external actions or changes to acceptance criteria without review. The threat surface is consistent with standard software engineering tasks and is handled through procedural constraints.
- [DATA_EXPOSURE_EXFILTRATION]: No sensitive data, API keys, or private filesystem paths are present. The documentation explicitly instructs developers to avoid storing secrets or machine-specific paths within the skill or its references.
- [COMMAND_EXECUTION]: The skill discusses command-line usage (e.g.,
grep,diff,docker exec) exclusively in the context of explaining testing traps and providing correct shell-scripting fixes. These are educational examples for the agent to use when generating test scripts, not commands intended for immediate or silent execution by the platform itself.
Audit Metadata