test-audit
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes several command-line tools such as
git,npx,mutmut,cargo, andgremlins. It uses the$ARGUMENTSvariable directly within these commands (e.g., in mutation tool execution), which could facilitate command injection if the underlying platform does not sanitize the input.\n- [DYNAMIC_EXECUTION]: As part of the mutation testing process, the skill creates modified versions of production source code (probes) and executes the test suite against these modified files. This involves the runtime modification and execution of code strings to verify test coverage effectiveness.\n- [INDIRECT_PROMPT_INJECTION]: The skill instructions specify that it can accept a 'change record' from a caller containing 'check commands'. These provided commands are used to identify the test runner and execute the suite, creating a surface for instruction injection.\n - Ingestion points: Data ingested from a 'change record' (including target, check commands, and baseline) as described in the Process/Inventory section.\n
- Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the change record data.\n
- Capability inventory: The skill possesses the capability to execute shell commands and modify local files using tools like
mv,git, and various package managers/test runners.\n - Sanitization: No sanitization, validation, or escaping mechanisms are described for the external commands provided in the change records.
Audit Metadata