honcho

Configure Honcho memory integration for Gengar with multi-profile peer isolation.

Updated May 4, 2026

Search Tags:#memory #context injection #cross-session #honcho #user modeling #agent profiles #recall modes

Ready to run

npx skills add https://github.com/JamesFincher/gengar --skill honcho-jamesfincher

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: honcho
Source: https://github.com/JamesFincher/gengar/tree/main/optional-skills/autonomous-ai-agents/honcho
Command: npx skills add https://github.com/JamesFincher/gengar --skill honcho-jamesfincher

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Honcho memory helps Gengar remember who the user is across sessions by maintaining a cross-session user model and profile-specific peer identities, reducing repetition and improving continuity.

Core Features & Use Cases

Configure Honcho for Gengar in cloud or self-hosted setups and verify resolved connections and peer info.
Tune multi-profile user modeling with isolated AI peers per Gengar profile while sharing a unified user representation.
Control recall and prompting behavior via recall modes (hybrid/context/tools), injection strategies (cold vs warm), and context budget enforcement.

Quick Start

Run the Honcho configuration wizard by executing gengar honcho setup and selecting the appropriate cloud or local option.

Frequently Asked Questions about honcho

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

FAQPage Schema

How do I configure cross-session memory for an AI agent?▼

Cross-session memory is configured by integrating Honcho with the Gengar agent, which maintains a unified user model and profile-specific peer identities across sessions to improve conversational continuity and reduce repetition.

Can I isolate AI peer identities across multiple user profiles?▼

Yes, multi-profile user modeling supports isolated AI peers per Gengar profile while sharing a unified user representation. This ensures peer isolation and prevents context bleed between distinct user profiles during cross-session recall.

How do I set up Honcho integration using a local or cloud configuration?▼

Run the Honcho configuration wizard via the setup command and select either the cloud or local self-hosted option. You must provide configuration parameters like the API key or base URL, peer identities, and workspace identities to verify the connection.

What recall modes are available for context injection?▼

Available recall modes include hybrid, context, and tools. These modes align session summaries, user representations, and AI peer cards into injected context, allowing you to control prompting behavior through cold or warm injection strategies.

How do I enforce context budget limits during cross-session recall?▼

Context budget enforcement is tuned alongside observation toggles, dialectic depth, cadence, and recall modes. Configuring these parameters ensures the injected context from session summaries and user representations stays within defined operational limits.

Do I need an API key to use Honcho memory integration?▼

Yes, Honcho configuration requires either an API key for cloud setups or a base URL for self-hosted local instances, along with peer and workspace identities, to successfully resolve connections and verify peer information.