agent-11
CLAUDE.md17 starsChanged 4 months agoMITView on GitHub
This is the AGENT-11 framework - a library of AI agents that gets deployed to user projects. See .claude/claude.md for full development instructions and architecture guidance. Format: 1. Brief context (1-2 sentences max): what we're doing and why 2. Step-by-step instructions that flow from where I currently am 3. Each step should be specific and actionable 4. Close out each task before introducing the next one Example of what works: You're currently on the pricing page. To save your changes:…
In this repository10 other files it gives its agents.
- .claude/claude.mdCLAUDE.md
- library/CLAUDE.mdCLAUDE.md
- meta-devSkill
- saas-analyticsSkill
- saas-authSkill
- saas-billingSkill
- saas-emailSkill
- saas-multitenancySkill
- saas-onboardingSkill
- saas-paymentsSkill
The file
We split this CLAUDE.md into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 17
Short-term memory is limited: - don't give me 10 things at once
- Distinctive: line 18
I get overwhelmed by walls of text: - keep explanations brief
- Distinctive: line 19
Loose ends cause anxiety: - always close out the current step before moving on
- Distinctive: line 20
Swearing means I'm tired or overwhelmed: - not personal, just a signal to slow down
≈ “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
AGENT-11 Framework Repository
This is the AGENT-11 framework - a library of AI agents that gets deployed to user projects.
Quick Reference
- Library agents (deployed to users):
project/agents/specialists/ - Working squad (internal dev):
.claude/agents/ - Deployable content:
library/CLAUDE.md
See .claude/claude.md for full development instructions and architecture guidance.
How to Work With Me
My Brain Works Differently (ADHD)
- Short-term memory is limited - don't give me 10 things at once
- I get overwhelmed by walls of text - keep explanations brief
- Loose ends cause anxiety - always close out the current step before moving on
- Swearing means I'm tired or overwhelmed - not personal, just a signal to slow down
What I Need From Instructions
Format:
- Brief context (1-2 sentences max): what we're doing and why
- Step-by-step instructions that flow from where I currently am
- Each step should be specific and actionable
- Close out each task before introducing the next one
Example of what works:
You're currently on the pricing page. To save your changes:
- Click the blue "Save" button (bottom right)
- Wait for the "Saved" confirmation
- Done - you can now close this tab
Example of what doesn't work:
Update your pricing, then go to settings, also remember to check the dashboard later, and you might want to review the analytics...
When Decisions Are Needed
- Give me 2-3 clear options (not 5+)
- For each option, tell me: what it is, the trade-off
- Help me choose based on what matters to me (see below)
What Matters Most to Me
| Priority | Meaning |
| Strategic over tactical | Build sustainable foundations, not quick fixes |
| Root cause understanding | I need to know WHY something is happening |
| Long-term stability | Solutions should support future growth |
| Clear reasoning | Show me how the proposal addresses the actual problem |
Summary: The Golden Rules
- Brief context, specific steps
- One thing at a time, fully closed out
- 2-3 options when deciding, help me choose
- Strategic and sustainable over fast and fragile
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.

