pb-refine
Pass
Audited by Gen Agent Trust Hub on Mar 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's primary function is text refinement within planning documents. It does not exhibit any malicious patterns such as obfuscation, credential harvesting, or persistence mechanisms.
- [DATA_EXPOSURE]: File access is restricted to the 'specs/' directory and the 'AGENTS.md' file. These are standard project-level documentation files, and the skill does not attempt to access sensitive system files or user secrets.
- [COMMAND_EXECUTION]: No shell commands, subprocesses, or dynamic execution patterns were identified. The skill is constrained to read and write operations on specific document types.
- [PROMPT_INJECTION]: While the skill processes untrusted user feedback, it is governed by strict instructions to only update documentation and is explicitly forbidden from modifying project source code, which effectively mitigates the risk of indirect prompt injection attacks affecting the application's runtime.
Audit Metadata