legacy-code-clean
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill explicitly instructs the agent to "proactively install" static analysis tools as needed. While it mentions a legitimate tool,
ts-prune, the instruction is open-ended ("including but not limited to"), which could result in the installation of unverified third-party packages. - [COMMAND_EXECUTION]: The skill performs shell-level operations, including package installation via scripts and the execution of static analysis tools on the codebase.
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to its core functionality of processing untrusted data.
- Ingestion points: The agent reads and analyzes source code, documentation (skill and dev docs), and git submission history within the project directory (
/workspace/gomtm). - Boundary markers: No specific delimiters or "ignore instructions" warnings are utilized to differentiate code logic from potentially malicious instructions embedded in comments or documentation.
- Capability inventory: The skill possesses significant capabilities, including the ability to delete source files, modify existing code, spawn sub-agents for tasks, and install new software packages.
- Sanitization: There is no evidence of content sanitization or validation of the files being read before they are used to influence the agent's decision-making regarding code deletion and system modifications.
Audit Metadata