testing-skill
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external source code and test files to provide guidance, which creates a surface for indirect prompt injection. Ingestion points: Project source code and test files are read during the analysis workflow; Boundary markers: No explicit delimiters or boundary markers are defined to isolate untrusted user content; Capability inventory: The skill includes an instruction to write a JSON sidecar file to the local file system; Sanitization: The skill lacks mechanisms to sanitize or validate the content of processed files before they are incorporated into the agent's reasoning process.
- [COMMAND_EXECUTION]: The 'Emit Outcome Sidecar' section instructs the agent to perform a file write operation to a hardcoded path (~/.claude/skill-analytics/last-outcome-testing-skill.json) to record execution metrics.
Audit Metadata