development-space
AGENTS.md1 starsChanged 12 days agoMITView on GitHub
Scope: how work moves from an idea to shipped code in this workshop, in any language. A tool-level AGENTS.md overrides this file. This file is the index. Read the one rule file that matches where you are, not all of them. Phase state comes from artifacts on disk: specs, spikes, ADRs, plans, worktrees, tests, code. A script reads rules/gates.yaml and reports which gates pass. There is no per-tool phases file to keep in sync. Durable decisions and open questions go…
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.
In this repository95 other files it gives its agents.
- AGENTS.mdAGENTS.md
- agent-architectureSkill
- agentic-evalSkill
- agent-owasp-complianceSkill
- agent-skill-stackSkill
- ai-readySkill
- ai-team-orchestrationSkill
- answer-me-with-htmlSkill
- anti-ui-slopSkill
- archifySkill
- autoresearchSkill
- batch-script-progressSkill
- cli-creatorSkill
- codebase-designSkill
- code-exemplars-blueprint-generatorSkill
- copilot-instructions-blueprint-generatorSkill
- create-agentsmdSkill
- create-architectural-decision-recordSkill
- create-github-action-workflow-specificationSkill
- create-github-issues-feature-from-implementation-planSkill
- create-readmeSkill
- create-specificationSkill
- create-technical-spikeSkill
- create-tldr-pageSkill
- docs-sync-auditSkill
- folder-structure-blueprint-generatorSkill
- gh-fix-ciSkill
- github-actions-hardeningSkill
- github-repo-careSkill
- impeccableSkill
- launchd-automationSkill
- mcp-security-auditSkill
- notion-cliSkill
- opencode-model-routerSkill
- playwright-cliSkill
- python-pypi-package-builderSkill
- refactorSkill
- simple-englishSkill
- swiftui-expert-skillSkill
- technical-documentationSkill
- workshop-driverSkill
- xerj-hubSkill
- xerjSkill
- anti-ui-slopSkill
- autoresearchSkill
- batch-script-progressSkill
- calling-frontend-from-tauri-rustSkill
- calling-rust-from-tauri-frontendSkill
- cli-creatorSkill
- codebase-designSkill
- code-exemplars-blueprint-generatorSkill
- configuring-tauri-appsSkill
- configuring-tauri-capabilitiesSkill
- configuring-tauri-cspSkill
- configuring-tauri-permissionsSkill
- configuring-tauri-scopesSkill
- copilot-instructions-blueprint-generatorSkill
- create-agentsmdSkill
- create-architectural-decision-recordSkill
- create-github-action-workflow-specificationSkill
- The first 60 shown.
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 24
When two rule files disagree, the phase rule wins over the index, and the index wins over nothing.
≈ “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
Workshop phases and skill routing
Scope: how work moves from an idea to shipped code in this workshop, in any language. A tool-level AGENTS.md overrides this file.
This file is the index. Read the one rule file that matches where you are, not all of them.
Which rule to read
| When | Read |
| Deciding small tool or platform/app | rules/tiers.md |
| Phase 1: vision, glossary, stack skills | rules/phase-1-vision.md |
| Phase 2: detect, prove, plan | rules/phase-2-plan.md |
| Phase 3: scaffold, build, verify | rules/phase-3-build.md |
| Recording the parts of a tool | rules/stack-profile.md |
| Reading a gate by script | rules/gates.yaml |
| Standard procedures (new tool, spike, ADR, ...) | driver/runbook.md |
| Known gaps in this routing | rules/known-gaps.md |
Precedence
- A tool's own
AGENTS.mdoverrides these rules. - A rule file overrides a default.
- When two rule files disagree, the phase rule wins over the index, and the index wins over nothing.
State is read, not filed
Phase state comes from artifacts on disk: specs, spikes, ADRs, plans, worktrees, tests, code. A script reads rules/gates.yaml and reports which gates pass.
There is no per-tool phases file to keep in sync. Durable decisions and open questions go in the tool's AGENTS.md, under Decisions and Open questions.
To read the current state and pick the next step, run the workshop-driver skill.
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.