breakdown-tasks
Pass
Audited by Gen Agent Trust Hub on Jul 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: Extensive review of the skill's markdown files, agent prompts, and support scripts confirmed that the tool operates purely on local project data. No malicious patterns, exfiltration attempts, or safety bypass mechanisms were identified.
- [PROMPT_INJECTION]: The skill defines clear roles for specialized agents and uses instructional guardrails to maintain output quality. It does not contain instructions that attempt to override the model's safety constraints or extract system prompts.
- [DATA_EXFILTRATION]: The skill manages its state within the local .forsvn directory. There are no network requests to external domains, and references to API keys in documentation are provided only as placeholders for user configuration.
- [REMOTE_CODE_EXECUTION]: No remote dependencies or scripts are fetched or executed. The provided TypeScript scripts are internal tools for managing the skill's metadata and local workflows, executing only in the local environment.
Audit Metadata