honcho

Configure and troubleshoot Honcho cross-session memory for Hermes agent profiles.

Updated Aug 22, 2026
One-click install
npx skills add https://github.com/vivekgoquest/hermes-agent-stable --skill honcho-vivekgoquest
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: honcho
Source: https://github.com/vivekgoquest/hermes-agent-stable/tree/main/optional-skills/autonomous-ai-agents/honcho
Command: npx skills add https://github.com/vivekgoquest/hermes-agent-stable --skill honcho-vivekgoquest

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires honcho-ai.

What problem does it solve? Setting up and maintaining persistent cross-session user memory in Hermes is complex: peers fail to sync, memory stops persisting, and multi-profile setups require careful configuration of observation, recall, and dialectic settings. ## Core Features & Use Cases - Setup & Verification: Guides cloud (app.honcho.dev) or self-hosted Honcho configuration via hermes memory setup honcho and validates with hermes honcho status. - Multi-Profile Peer Management: Creates per-profile AI peers with hermes profile create --clone and backfills existing profiles using hermes honcho sync. - Dialectic Tuning: Explains the three orthogonal knobs — cadence, depth, and reasoning level — plus context budgets, recall modes, and session strategies. - Use Case: When memory stops persisting across sessions, run hermes honcho status to verify saveMessages and writeFrequency settings, then adjust the config as directed. ## Quick Start Ask the agent to set up Honcho memory for Hermes and verify the connection status.

Frequently Asked Questions about honcho

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I set up Honcho memory for Hermes?▼

Run `hermes memory setup honcho` and choose cloud (paste an API key from app.honcho.dev) or local (enter a base URL such as http://localhost:8000). Verify the configuration with `hermes honcho status`, which shows resolved config, a connection test, and peer info.

How do I give each Hermes profile its own Honcho peer?▼

Create profiles with `hermes profile create <name> --clone`, which adds a host block in honcho.json, sets the aiPeer to the profile name, and inherits shared settings. For existing profiles, run `hermes honcho sync` to backfill missing host blocks.

Why is Honcho memory not persisting across sessions?▼

Check `hermes honcho status` and verify `saveMessages` is true and `writeFrequency` is not set to `session`, which only writes on exit. Also confirm `memory.provider: honcho` is present in ~/.hermes/config.yaml.

What is the difference between hybrid, context, and tools recall modes?▼

Hybrid (default) auto-injects context and exposes the five Honcho tools. Context mode auto-injects but hides tools for minimal token cost. Tools mode disables auto-injection so the agent controls all memory access explicitly.

When should I use honcho_reasoning versus honcho_search?▼

Use honcho_search for fast, no-LLM retrieval of specific past facts as raw excerpts. Use honcho_reasoning only when you need a synthesized answer from Honcho's dialectic engine, since it triggers an LLM call with medium-to-high cost.