generate-testcase
Pass
Audited by Gen Agent Trust Hub on Jun 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from git commits and user arguments to generate and execute code.
- Ingestion points: User input via
$ARGUMENTSand file content/history from git commits (SKILL.md). - Boundary markers: No delimiters or isolation instructions are present to separate untrusted commit data from the agent's logic.
- Capability inventory: The skill is designed to build and run test cases, which involves local command execution (SKILL.md).
- Sanitization: The skill does not perform any validation or sanitization of the content extracted from commits before using it to generate executable test functions.
- [COMMAND_EXECUTION]: The skill requires the agent to perform local command execution to build the project, run tests, and generate coverage reports using tools like fastcov and lcov (SKILL.md).
Audit Metadata