wm

Manage AI session memory with wm CLI commands like dive, compile, and distill.

3|Updated Jan 7, 2026
One-click install
npx skills add https://github.com/open-horizon-labs/bottle --skill wm
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: wm
Source: https://github.com/open-horizon-labs/bottle/tree/main/codex-skill/wm
Command: npx skills add https://github.com/open-horizon-labs/bottle --skill wm

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

wm provides a structured external memory for AI sessions, enabling you to capture tacit knowledge from work, recall relevant context on demand, and distill learnings into actionable guidance.

Core Features & Use Cases

  • Dive prep & session context: Prepare and reference context before and during AI sessions.
  • Context compilation: Gather past decisions and knowledge to assemble a coherent working context.
  • Learning distillation: Extract learnings after sessions to improve future work.

Quick Start

Install the wm CLI, initialize your project and create a first dive:

  1. Install: brew tap open-horizon-labs/homebrew-tap && brew install wm
  2. Initialize: wm init
  3. Start a session: wm dive plan

Frequently Asked Questions about wm

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

FAQPage Schema
How do I maintain context across multiple AI sessions?

You maintain context across AI sessions by using an external working memory tool to capture tacit knowledge, compile past decisions, and prepare context before starting new iterative work.

What is the best way to prepare context before starting an AI session?

Preparing context for an AI session involves using a dive prep command that loads relevant past knowledge and working memory into your current environment.

How do I distill learnings from AI sessions into actionable guidance?

Distilling learnings from AI sessions requires applying a distillation command that extracts tacit knowledge from your work and stores it as actionable guidance for future runs.

Do I need to install a CLI to manage working memory for AI sessions?

Yes, managing working memory for AI sessions requires installing a dedicated CLI tool, which exposes commands to initialize projects, compile context, and manage session prep.

Can I compile past decisions and knowledge into a coherent working context?

You can compile past decisions and knowledge into a coherent working context by using a context compilation command to assemble relevant information for your AI sessions.

When do I need external memory management for iterative AI work?

External memory management is needed for iterative AI work when sessions require prep, context loading, and learning distillation across multiple runs to maintain continuity.