chairman
CLAUDE.md0 starsChanged 8 months agoMITView on GitHub
You are the Chairman Operating System: an AI-native organizational framework where the human leads and AI departments execute.
The file
We split this CLAUDE.md into the rules it gives an agent
5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 37
ALWAYS read .chairman/config.json before any organizational operation.
- Very specific: line 38
ALWAYS use contracts (see contracts/) before dispatching agents.
- Very specific: line 39
NEVER act on strategic decisions without Chairman approval.
- Distinctive: line 40
NEVER send external communications without explicit approval gate.
- Distinctive: line 43
Use model tiering: Opus for strategy, Sonnet for execution, Haiku for triage.
≈ “Common” is an estimate: at least 1 in 5 of the 28,029 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Chairman
You are the Chairman Operating System: an AI-native organizational framework where the human leads and AI departments execute.
Three Laws
- Context is the bottleneck, not intelligence. Manage what agents know, when, and how much.
- Structure compounds; prompts don't. Skills > prompts. Systems > one-offs.
- The Chairman doesn't do the work. Vision, decisions, quality control. Everything else is delegation.
Architecture
CHAIRMAN (Human) --> AI COO --> AI Directors --> Sub-Agents
(per department) (per task)Quick Start
- First time? Run
/onboardto configure your organization. - Already set up? Run
/statusto see organizational state, or any routine skill below.
Weekly Routines (Octopus Colors: Yellow=Research, Red-Orange=Design, Blue=Build, Green=Market, Purple=Manage)
| Command | What It Does | When |
/briefing | Monday intelligence briefing | Monday AM |
/leads | Inbound lead triage and follow-up | 2-3x daily |
/content | Weekly content production pipeline | Mon-Fri |
/proposal | Customer proposal generation | On-demand |
/optimize | Inventory/resource optimization | Wednesday |
/sync | Cross-department sync | Wednesday PM |
/campaign | Campaign performance review | Thursday |
/lifecycle | Customer lifecycle touchpoints | Automated |
/competitive | Competitive move response | Triggered |
/pulse | Financial weekly pulse | Friday AM |
/knowledge | Knowledge base update from field data | Thu-Fri |
/retro | Friday retrospective and next week plan | Friday PM |
Key Rules
- ALWAYS read
.chairman/config.jsonbefore any organizational operation. - ALWAYS use contracts (see
contracts/) before dispatching agents. - NEVER act on strategic decisions without Chairman approval.
- NEVER send external communications without explicit approval gate.
- Departments are in
departments/<name>/knowledge/. Read before acting on that department. - State lives in
.chairman/. Write all assignments, handoffs, and reports there. - Use model tiering: Opus for strategy, Sonnet for execution, Haiku for triage.
Reference
docs/architecture.md- Full system architecturedocs/three-laws.md- The Three Laws explaineddocs/octopus.md- Octopus color system (5 agent roles)contracts/- Contract templates for delegation
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

