skills.rest
villadalmine
villadalmineCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

honcho

Deploy a persistent, workspace-isolated memory platform for AI agents.

1|Updated Apr 3, 2026
Search Tags:#postgresql#redis#jwt#kubernetes#memory#honcho
One-click install
npx skills add https://github.com/villadalmine/infra --skill honcho-villadalmine
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: honcho
Source: https://github.com/villadalmine/infra/tree/main/skills/honcho
Command: npx skills add https://github.com/villadalmine/infra --skill honcho-villadalmine

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Honcho solves the problem of durable, isolated memory for AI agents by keeping workspaces, sessions, peers, and chat history persistent across restarts while preventing one consumer from reading another consumer’s state.

Core Features & Use Cases

  • Workspace-scoped memory: Use JWT HS256 keys to isolate memory access for agents like OpenClaw and Hermes.
  • Production-grade deployment: Run the full stack with Postgres, Redis, API, and worker components plus Alembic migrations handled automatically in init containers.
  • Cluster integration: Connect the service through Kubernetes, Cilium-safe NetworkPolicies, and a stable Service-to-pod port mapping for reliable agent access.
  • Use case: Deploy a long-lived assistant that remembers prior conversations, stores derived context, and serves multiple internal consumers without cross-workspace leakage.

Quick Start

Deploy Honcho for my cluster with workspace-isolated JWT auth, automatic database migrations, and the correct Kubernetes networking setup.

Frequently Asked Questions about honcho

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

FAQPage Schema
How do I deploy persistent memory for AI agents in Kubernetes?▼

Deploy persistent memory for AI agents in Kubernetes by running a stack with Postgres, Redis, API, and worker components, using Alembic init-containers for automatic database migrations and JWT HS256 keys to isolate workspace state across restarts.

How does JWT HS256 isolate AI agent memory across workspaces?▼

JWT HS256 isolates AI agent memory by scoping access keys to specific workspaces, ensuring agents like OpenClaw and Hermes maintain separate sessions, peers, and chat history without cross-workspace data leakage.

Do I need Postgres and Redis to run workspace-isolated memory for agents?▼

Yes, Postgres and Redis are required dependencies to run workspace-isolated memory for agents, providing the durable storage and caching backbone for chat history, sessions, and long-running workspace state.

Why do Cilium egress rules target pod port 8000 instead of service port 80?▼

Cilium egress rules target pod port 8000 instead of service port 80 to ensure reliable agent access through stable Service-to-pod port mapping within the Kubernetes cluster networking setup.

What is the best way to prevent cross-consumer state leakage in AI memory platforms?▼

The best way to prevent cross-consumer state leakage in AI memory platforms is using workspace-scoped memory with JWT HS256 authentication, keeping sessions, peers, and chat history strictly isolated per consumer.

Related Skills

Explore More Software Engineering Skills→
rohitg00rohitg00
944

k8s-cilium

Observe and troubleshoot Cilium networking in Kubernetes clusters.

Community
Advanced
marmikcfcmarmikcfc

pepper-state

Read, write, append, and list persistent state via the Pepper Cloud API.

Community
Intermediate
altrupetsaltrupets

Network Policies & Security

Manage Kubernetes network policies to control pod and namespace traffic flows.

Official
Advanced
joonlabjoonlab
8

langgraph-persistence

Manage durable state persistence and long-term memory for LangGraph agents.

Community
Advanced