honcho

Configure cross-session user modeling and dialectic memory for Hermes agents.

1|Updated May 12, 2026
One-click install
npx skills add https://github.com/projectedanx/hermes-agent --skill honcho-projectedanx
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: honcho
Source: https://github.com/projectedanx/hermes-agent/tree/main/optional-skills/autonomous-ai-agents/honcho
Command: npx skills add https://github.com/projectedanx/hermes-agent --skill honcho-projectedanx

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires honcho-ai.

What problem does it solve?

Honcho solves the problem of AI agents forgetting user preferences, facts, and interaction patterns across different sessions, ensuring a consistent and deepening model of the user over time.

Core Features & Use Cases

  • Cross-Session User Modeling: Accumulates facts and preferences to build a persistent identity of the user.
  • Dialectic Reasoning: Uses a multi-pass reasoning engine to synthesize complex insights about user goals and patterns.
  • Multi-Profile Isolation: Allows different agent profiles to maintain independent AI identities while sharing a unified workspace.
  • Use Case: Use this to maintain context across long-term projects, ensuring the agent remembers your coding style, preferred tools, and past project decisions without needing to be reminded.

Quick Start

Run the honcho setup command to initialize your memory provider and configure your preferred synchronization settings.

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 persistent user modeling in AI agents?

Cross-session user modeling is enabled by running the honcho setup command to initialize a memory provider, which accumulates facts and preferences to build a persistent user identity across different sessions.

What is dialectic memory and how does it work for autonomous fact-gathering?

Dialectic memory uses a multi-pass reasoning engine to synthesize complex insights about user goals and patterns. It enables autonomous fact-gathering by bidirectionally observing user-AI interactions to continuously update the persistent user model.

Can I maintain independent agent profiles while sharing a unified workspace?

Yes, multi-profile isolation allows different agent profiles to maintain independent AI identities while sharing a unified workspace. This ensures each profile retains its specific context and behavioral patterns without cross-contamination.

Does honcho work with Hermes agents for context budget enforcement?

Yes, honcho configures and manages dialectic memory specifically for Hermes agents. It satisfies requirements for context budget enforcement by summarizing sessions and enforcing limits on persistent context usage.

Do I need the honcho-ai dependency to set up session summarization?

Yes, the honcho-ai dependency is required to configure synchronization settings and enable session summarization. It provides the underlying infrastructure for cross-session user modeling and persistent context management.

When should I not use cross-session memory for user modeling?

Cross-session memory is not ideal for single-session or stateless interactions where persistent context adds unnecessary overhead. It is best suited for long-term projects requiring accumulated knowledge of user preferences and past decisions.