mutation-check
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands such as
cargo mutants --list -f <file>andcargo mutants -f src/<file>.rs -- --libto perform mutation testing. These commands run locally within the developer environment. - [INDIRECT_PROMPT_INJECTION]: The skill analyzes external data that could theoretically be influenced by an attacker.
- Ingestion points: The skill reads source code diffs, the
.cargo/mutants.tomlconfiguration file, and the triage filemutants.out/missed.txtgenerated by external tools. - Boundary markers: None identified in the instructions for delimiting the file content from the agent's instructions.
- Capability inventory: The agent has the capability to execute shell commands (via
cargo mutants) and modify local files (e.g., updating.cargo/mutants.toml). - Sanitization: The skill does not explicitly mention sanitizing or validating the contents of the diffs or tool output before processing.
Audit Metadata