amplifier-converge
AGENTS.md1 starsChanged yesterdayMITView on GitHub
This codebase is yours: what it's for and how it's built are your call, and you'll keep building on it. This repository develops the Amplifier Converge methodology. Start with README.md, docs/VISION.md, and the applicable contracts. For work applying the method elsewhere, use guides/agent-guide.md; for a guidance review, use guides/reviewer-guide.md. Keep the shared promises portable across people and agent systems. Put practical application in the guides, with links to the governing contracts. Keep the methodology independent of a particular runtime, database,…
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,439 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
The file
We split this AGENTS.md into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
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 11
When editing a commitment, consider the behavior and audience it affects.
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.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
Working in this repository
This codebase is yours: what it's for and how it's built are your call, and you'll keep building on it.
This repository develops the Amplifier Converge methodology. Start with README.md, docs/VISION.md, and the applicable contracts. For work applying the method elsewhere, use guides/agent-guide.md; for a guidance review, use guides/reviewer-guide.md.
Keep the shared promises portable across people and agent systems. Put practical application in the guides, with links to the governing contracts. Keep the methodology independent of a particular runtime, database, application, or agent framework. Project contracts may deliberately commit to technologies, architecture, interfaces, and detailed experience behavior.
Use plain language, concrete examples, and just enough detail to act or judge. Keep technical and experience details when they define something others must rely on; brevity must not make a commitment ambiguous. Preserve one authoritative home for each promise. Independent decisions within delegated authority are the normal mode; do not add approval gates for routine implementation or ordinary editorial work. Proactively expose consequential gaps in direction before dependent implementation and help resolve them; check clarity for the next outcome, not completeness of the whole project.
When editing a commitment, consider the behavior and audience it affects. Keep implementation detail and review evidence out of the vision. Describe the methodology directly; do not add migration notes, references to previous implementations, or a changelog.
For documentation changes, inspect the complete changed passages, check local links and references, and walk through a relevant scenario. Do not claim a documentation review proves live execution. Do not provision an application or install dependencies just to review prose.
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.