protocol-writer
Pass
Audited by Gen Agent Trust Hub on Apr 15, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No malicious instructions, safety bypasses, or attempts to extract system prompts were detected in the skill instructions.
- [DATA_EXFILTRATION]: The skill only interacts with local project files (STATUS.md, GOAL.md, DECISIONS.md) and writes to a local output folder. No sensitive system paths are accessed, and no network operations are present.
- [REMOTE_CODE_EXECUTION]: The skill does not download or execute any external scripts or packages.
- [COMMAND_EXECUTION]: There are no shell commands, subprocess invocations, or use of the dynamic context injection syntax (!command).
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from local project files. However, its capabilities are restricted to generating a Markdown document, which does not pose a security risk as there are no executable downstream tools or network exfiltration paths associated with this process.
Audit Metadata