alexai-mcp/hermes-ccc/honcho
Claude Code and agent skill by alexai-mcp/hermes-ccc
- community
Checking that the SKILL.md file is still in the repository.
AI-native cross-session user modeling — builds a persistent, growing model of who you are, your expertise, preferences, and working style across all Claude sessions.
Install
npx vskill@latest install alexai-mcp/hermes-ccc/honchoSKILL.md
Honcho — Cross-Session User Modeling
"The agent that grows with you." Honcho builds a persistent model of who you are across every session — your expertise, communication style, recurring topics, and working preferences.
Philosophy
Unlike simple memory (saving facts), Honcho models the user as a person — updating its understanding through dialectic reasoning: observing patterns, forming hypotheses about your preferences, and refining them over time.
This is the core of what makes Hermes Agent feel like it "knows" you.
hermes-CCC Implementation
In hermes-CCC, Honcho runs as a structured user profile in the memory system. Full Honcho (honcho.dev) adds a cloud backend with dialectic reasoning.
User Profile Structure
Profile stored at: ~/.claude/projects/*/memory/user_honcho_profile.md
Commands
/honcho profile
Display the current user model. Claude reads the profile and summarizes key facts about how it understands you.
/honcho update
After a session, analyze what was discussed and update the profile:
- New expertise demonstrated?
- New tools or projects mentioned?
- Communication preferences revealed?
- New recurring topics?
Then write updates to user_honcho_profile.md.
/honcho calibrate
Run a quick 5-question calibration:
- What's your primary role?
- What are your strongest technical areas?
- How do you prefer explanations? (terse/detailed)
- What projects are you currently working on?
- Any specific preferences for how I should behave?
/honcho reset
Clear the user model and start fresh.
/honcho export
Export profile as JSON for backup or transfer.
How Claude Uses the Profile
When Honcho profile is loaded, Claude should:
- Skip explanations of tools/concepts the user knows deeply
- Use preferred verbosity level in all responses
- Reference current projects when suggesting approaches
- Adapt language mix (Korean/English as preferred)
- Assume expertise level from profile when writing code
- Prioritize working patterns (e.g., "delegate to Codex")
Full Honcho Integration (Optional)
For cloud-backed dialectic reasoning:
See docs.honcho.dev for full API reference.
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/honcho/SKILL.md
Popularity
- Repository stars
- 134
- Repository forks
- 23
- 7-day trend score
- 0
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/alexai-mcp/hermes-ccc/honcho)