honcho

Configure Honcho memory in Hermes for persistent cross-session user modeling.

Updated May 26, 2026
One-click install
npx skills add https://github.com/ruiyangruiyi/hermes-agent --skill honcho-ruiyangruiyi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: honcho
Source: https://github.com/ruiyangruiyi/hermes-agent/tree/main/optional-skills/autonomous-ai-agents/honcho
Command: npx skills add https://github.com/ruiyangruiyi/hermes-agent --skill honcho-ruiyangruiyi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Honcho helps Hermes maintain accurate, cross-session user modeling—so your agent’s recommendations, preferences, and continuity improve over time instead of resetting each chat.

Core Features & Use Cases

  • Cross-session user modeling: Builds a persistent representation of the user and injects it into Hermes prompts for continuity.
  • Multi-profile peer isolation: Creates separate AI peers per Hermes profile while sharing a unified user view.
  • Controlled recall modes: Supports hybrid, context-only, and tool-driven memory access with context budget enforcement.
  • Dialected reasoning & session summaries: Tunes cadence, depth, and reasoning level for Honcho’s dialectic engine, with session summary injection to reduce token costs.
  • Observation & session strategies: Configures what gets observed per peer and how sessions are scoped (per-directory, per-repo, per-session, global).

Quick Start

Configure Honcho memory for Hermes by running: hermes honcho setup

Frequently Asked Questions about honcho

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

FAQPage Schema
How do I enable cross-session memory for user modeling in an agent?

Cross-session memory is enabled by configuring Honcho inside Hermes using a valid honcho.json file and running the setup command to build a persistent user representation. This injects user context into prompts for continuous multi-session modeling.

What is the best way to isolate memory across multiple agent profiles?

Multi-profile peer isolation is achieved by creating separate AI peers per Hermes profile while sharing a unified user view. This scoped observation behavior ensures profile-specific memory separation without fragmenting the core user modeling context.

How do I configure recall modes and context budgets for memory tools?

Recall modes and context budgets are configured in the honcho.json file by selecting hybrid, context-only, or tool-driven memory access. This setup enforces specific context limits to manage token costs during cross-session retrieval.

Can I use dialectic reasoning and session summaries to reduce token costs?

Dialectic reasoning and session summaries reduce token costs by tuning the reasoning cadence and injecting summarized session data into prompts. This approach limits unnecessary context expansion while preserving cross-session modeling accuracy.

Why is my Honcho memory sync not working between sessions?

Memory sync issues often stem from an invalid Honcho provider setup or incorrect honcho.json configuration. Troubleshooting requires verifying the recallMode selection and ensuring proper use of Honcho tools for context and profile management.