antigravity-fullstack-hq
AGENTS.md30 starsChanged 52 days agoMITView on GitHub
This repository is a configuration and documentation kit. It has no application runtime, package manager, or production service to deploy. - core/rules/common.md contains host-neutral engineering policy. - adapters/<host>/rules.md contains only host-specific integration guidance; adapters/project/rules.md is the adapter for repository (project) installs. - agents/ contains canonical Markdown agent bodies. - skills/ contains canonical Agent Skills. - workflows/ contains canonical workflow bodies and the Antigravity legacy bridge. - plugin.json packages the existing canonical skills for plugin-aware hosts. - .claude-plugin/ and hooks/hooks.json expose…
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 31,843 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 repository30 other files it gives its agents.
- claude/CLAUDE.mdCLAUDE.md
- llms.txtllms.txt
- api-design-patternsSkill
- auth-patternsSkill
- backend-dev-guidelinesSkill
- brainstormingSkill
- code-review-patternsSkill
- deployment-guideSkill
- docker-patternsSkill
- docx-officialSkill
- frontend-designSkill
- github-actionsSkill
- git-workflowSkill
- nestjs-patternsSkill
- nextjs-app-routerSkill
- pdf-officialSkill
- pptx-officialSkill
- prisma-workflowSkill
- prompt-engineeringSkill
- react-best-practicesSkill
- security-checklistSkill
- skill-creatorSkill
- software-architectureSkill
- systematic-debuggingSkill
- tailwind-patternsSkill
- test-driven-developmentSkill
- typescript-patternsSkill
- webapp-testingSkill
- web-design-guidelinesSkill
- xlsx-officialSkill
The file
We split this AGENTS.md into the rules it gives an agent
6 rules. About 0 are common, 4 are distinctive, and 2 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.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 22
Do not create separate copies of shared agent or skill instructions for each host.
- Distinctive: line 28
Run the platform-native validator before proposing a release:
- Very specific: line 42
Keep the existing top-level agents/, skills/, and workflows/ paths for compatibility.
- Very specific: line 44
Update README, setup/customization docs, and CHANGELOG.md when behavior or supported hosts change.
- Distinctive: line 46
Do not put host-specific tool names, paths, or commands in canonical agent and skill bodies unless the example is explicitly about that …
- Distinctive: line 48
Prefer additive migration bridges over deleting legacy installation paths.
≈ “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
Full Stack HQ repository instructions
This repository is a configuration and documentation kit. It has no application runtime, package manager, or production service to deploy.
Source of truth
core/rules/common.mdcontains host-neutral engineering policy.adapters/<host>/rules.mdcontains only host-specific integration guidance;adapters/project/rules.mdis the adapter for repository (project) installs.agents/contains canonical Markdown agent bodies.skills/contains canonical Agent Skills.workflows/contains canonical workflow bodies and the Antigravity legacy bridge.plugin.jsonpackages the existing canonical skills for plugin-aware hosts..claude-plugin/andhooks/hooks.jsonexpose the canonical agents, skills, workflows, and shared rules as a Claude Code plugin marketplace.bootstrap.*fetch a temporary checkout and run the matching installer.llms.txtanddocs/FAQ.mdsummarize the project for AI assistants and people; keep their counts, commands, and links in sync with the source.scripts/build-adapters.*renders Codex TOML agents and workflow-to-skill adapters.
Do not create separate copies of shared agent or skill instructions for each host. Put shared behavior behind the smallest useful interface and keep host differences in an adapter.
Verification
Run the platform-native validator before proposing a release:
PowerShell: .\scripts\validate.ps1
Bash: bash scripts/validate.shThe validator must pass on Windows and Linux. Installers should be tested with their dry-run option and with the isolated smoke tests. The doctor is read-only; it may report missing host installations as warnings. These checks must not push, create branches, or deploy.
Change discipline
- Keep the existing top-level
agents/,skills/, andworkflows/paths for compatibility. - Update README, setup/customization docs, and
CHANGELOG.mdwhen behavior or supported hosts change. - Do not put host-specific tool names, paths, or commands in canonical agent and skill bodies unless the example is explicitly about that host.
- Prefer additive migration bridges over deleting legacy installation paths.
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.