adlc-toolkit
AGENTS.md173 starsChanged 12 days agoMITView on GitHub
This project runs spec-driven development with the ADLC toolkit (Agentic Development Life Cycle). Every agent harness this project uses reads this file from the project root; settings that only one harness understands live in that harness's own files. Requirements come before code. /spec drafts a requirement, /validate checks it, /architect designs it and breaks it into tasks, then the tasks are implemented, self-reviewed with /reflect, reviewed, and closed out with /wrapup. /proceed REQ-xxx runs the whole pipeline for one requirement,…
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 31,774 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 38
Read these before changing the project:
≈ “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
AGENTS.md
This project runs spec-driven development with the ADLC toolkit (Agentic Development Life Cycle). Every agent harness this project uses reads this file from the project root; settings that only one harness understands live in that harness's own files.
Workflow
Requirements come before code. /spec drafts a requirement, /validate checks
it, /architect designs it and breaks it into tasks, then the tasks are
implemented, self-reviewed with /reflect, reviewed, and closed out with
/wrapup. /proceed REQ-xxx runs the whole pipeline for one requirement, and
/sprint runs several in parallel.
Skills
/adversary/analyze/architect/bugfix/canary/init/manifest/optimize/proceed/reflect/review/spec/sprint/status/template-drift/validate/wrapup
Project context
Read these before changing the project:
.adlc/ETHOS.md— the principles every skill follows.adlc/context/project-overview.md— what this project is and who uses it.adlc/context/architecture.md— how it is built.adlc/context/conventions.md— how code is written here.adlc/specs/— requirements, with their architecture and tasks.adlc/knowledge/— lessons learned and validated assumptions
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.