Skip to content
Get packs

OpenRig Operating Model

Teaches an agent where knowledge, notes and specs belong across a fleet of agent seats.

Source Download skill
Repository
mvschwarz/openrig
Stars
8,553
Licence
Apache-2.0
Files
11 files

What it does

It gives an AI agent a structured way to decide where a lesson, note, document or convention should live inside an OpenRig multi-agent setup. The agent learns to find context by reading one consistently named file at each level of a folder tree, from its own position up to the fleet, and to avoid duplicating knowledge. It also ships starter templates and small shell and Python helpers that create missing files without overwriting existing ones and check whether a trace is due.

Great for

  • Developers building or operating multi-agent OpenRig topologies with many seats, rigs and pods
  • Teams that keep agent memory and project specs as Markdown files and want consistent placement rules
  • Agent authors who want an example of a path-based context convention

Skip it if

  • Single-chat users who do not run multi-agent setups; the concepts assume OpenRig's topology and work trees and add little value alone
  • Projects that cannot have agents create or edit files in their repositories, since the scaffold scripts write files into folders
  • Anyone expecting it to work without the OpenRig product, as the skill references library addresses and plugins that may not exist in your environment

Try asking

  • “I learned something about deployment today. Which file should it go in, LEARNED.md or a new note?”
  • “Should this be a skill, a note, or a chain file? Help me decide.”
  • “Scaffold the topology chain files for these seat directories under my OpenRig root.”
  • “Before I create a new docs folder, check what already exists at each level of the tree.”
Step-by-step guide: how to use OpenRig Operating ModelInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 11 original files, the Apache-2.0 licence and an attribution note.

    openrig-operating-model.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip openrig-operating-model.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip openrig-operating-model.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Run the scaffold script in a test copy first, since it creates files in your tree even though it never overwrites existing ones.
  • Treat LEARNED.md and NOTES.md as lived files written in the agent's own words; keep generated renders separate from them.
  • Read the SKILL.md grid tables before asking an agent to place new knowledge, since the rules are specific about which file is authored by whom.

What's inside

  • scripts/compose.py
  • scripts/scaffold.sh
  • scripts/trace-due.sh
  • scripts/trace-stamp.sh
  • scripts/trunk-diff.sh
  • SKILL.md
  • templates/LEARNED.md
  • templates/NOTES.md
  • templates/PLAYBOOK.md
  • templates/SOP.md
  • templates/SPEC.md

Redistributed unmodified from mvschwarz/openrig under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.