honcho
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a personalization and memory system by maintaining a structured user profile at
~/.claude/projects/*/memory/user_honcho_profile.md. This is a standard practice for persistent agent memory. - [EXTERNAL_DOWNLOADS]: The documentation includes an optional integration section that references the official
honcho-aiPython package and thehoncho.devservice. These are legitimate resources provided for the skill's stated purpose of advanced user modeling. - [COMMAND_EXECUTION]: The custom commands (e.g.,
/honcho update,/honcho calibrate) are high-level instructions for the agent to manage its internal profile data and do not involve execution of arbitrary or dangerous system commands.
Audit Metadata