agent
AGENTS.md0 starsChanged 36 days agoMITView on GitHub
This public repository is the canonical source for the unified Vitae plugin: hosted connector, one platform operating skill, and nine recruiting workflows. Keep product implementation and customer information out of it. The old vitaedotai/skills repository is a historical snapshot, not a second active source.
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 repository11 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- candidate-outreachSkill
- candidate-presentationSkill
- candidate-screeningSkill
- hiring-scorecardSkill
- interview-invitationSkill
- interview-kitSkill
- job-intakeSkill
- recruitment-workflowSkill
- sourcing-strategySkill
- vitaeSkill
The file
We split this AGENTS.md into the rules it gives an agent
5 rules. About 0 are common, 1 is distinctive, and 4 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.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 6
Use the actual live tool schema.
- Very specific: line 7
Keep every connector pointed at https://mcp.vitae.ai/mcp.
- Very specific: line 8
Run python3 scripts/validate.py and the regression suite on an authorized verification host.
- Distinctive: line 10
Use scoped branches and PRs.
- Very specific: line 12
Keep all ten skill folders under skills/, with local resources inside each folder so individual installation remains self-contained.
≈ “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
Maintaining the Vitae agent package
This public repository is the canonical source for the unified Vitae plugin: hosted connector, one platform operating skill, and nine recruiting workflows. Keep product implementation and customer information out of it. The old vitaedotai/skills repository is a historical snapshot, not a second active source.
plugin.jsonis the release-version authority. Keep native manifests, marketplace metadata,server.json, catalog.json, every skill version, and CHANGELOG aligned.- Use the actual live tool schema. Refresh the public snapshot with
python3 scripts/refresh_tools.py. Never invent tools or publish authenticated responses. - Keep every connector pointed at
https://mcp.vitae.ai/mcp. OAuth is the default; never embed credentials. - Run
python3 scripts/validate.pyand the regression suite on an authorized verification host. Follow the user's host restrictions. - A green packaging check is not authenticated client acceptance. Record actual client results in
docs/compatibility.md; never claim a marketplace listing merely because its manifest exists. - Use scoped branches and PRs. Do not weaken Vitae's server approval controls through skill wording.
- Keep all ten skill folders under
skills/, with local resources inside each folder so individual installation remains self-contained. Validate catalog parity and build the portable upload ZIP withpython3 scripts/package.py. - Required current-head CI governs merge. Separate reviewers require an explicit user request.
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.