honey-compress
Pass
Audited by Gen Agent Trust Hub on Jun 26, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill performs local file system operations to create backups and update text files. This is restricted to the workspace and aligned with its stated purpose.
- [INDIRECT_PROMPT_INJECTION]: The skill processes existing project files which could contain untrusted data. It lacks explicit boundary markers for input data but mitigates risk through specific instructions to preserve the meaning and logic of the original content.
- Ingestion points: Project files (e.g., CLAUDE.md, AGENTS.md, notes).
- Boundary markers: Absent.
- Capability inventory: File system read and write access for backup and update operations.
- Sanitization: None performed on the file content before rewriting.
Audit Metadata