test-audit
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is configured to execute shell commands and scripts defined in the host repository's configuration files.
- Evidence: SKILL.md instructs the agent to 'Run tests through the repository's own commands (package.json scripts, Makefile, CI workflows)'.
- Evidence: references/campaign.md specifies the use of shell utilities like 'git rev-parse' and 'cmp'.
- [INDIRECT_PROMPT_INJECTION]: The skill ingest potentially untrusted data from the local repository to drive file modifications and deletions.
- Ingestion points: Analyzes source code, test files, git history, and CI configurations (audit.md Step 2, campaign.md Step 3).
- Boundary markers: Instructions do not include specific delimiters or directives to ignore instructions that might be embedded in the analyzed code or comments.
- Capability inventory: The skill has the capability to delete or edit files (audit.md Step 3) and execute local scripts (SKILL.md).
- Sanitization: No validation or sanitization of codebase content is mentioned before it influences the agent's logic.
Audit Metadata