calven-mcp
AGENTS.md3 starsChanged 39 days agoMITView on GitHub
This repository is the public distribution surface for Calven's hosted MCP server. It does not own or contain the production server implementation. Keep mcp.json, .mcp.json, examples, documentation, and skills aligned with that endpoint. Run npm ci --prefix scripts once, then npm run validate --prefix scripts after every change. Use the bare brand name Calven. Use calven.ai only when referring to the domain. Keep prose direct and specific. Avoid unsupported claims and promotional filler. Terminology: "sign in" (not "authenticate"); "Client ID…
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.
In this repository13 other files it gives its agents.
- battlecardSkill
- customer-evidenceSkill
- deal-prepSkill
- gtm-overviewSkill
- icp-checkSkill
- icp-pipeline-reviewSkill
- market-researchSkill
- on-message-draftSkill
- persona-copy-reviewSkill
- product-intelligenceSkill
- quality-reviewSkill
- startSkill
- win-loss-reviewSkill
The file
We split this AGENTS.md into the rules it gives an agent
4 rules. About 1 is common, 1 is distinctive, and 2 are very specific to this project.
≈1
Common · estimate
Many files like this have a rule of this kind.
≈1
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
- could be automatic: a setting could enforce it
- Very specific: line 14
Keep mcp.json, .mcp.json, examples, documentation, and skills aligned with that endpoint.
- Common: line 18
Never add a credential, tenant identifier, customer name, or private server source.
- Distinctive: line 20
Do not add or rename a documented tool without verifying the production catalog.
- Very specific: line 26
Use the bare brand name Calven.
≈ “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
Repository instructions
This repository is the public distribution surface for Calven's hosted MCP server. It does not own or contain the production server implementation.
Sources of truth
server.jsonis the Official MCP Registry descriptor.plugin.json,mcp.json, andskills/are the portable Agent Plugin.docs/tool-catalog.mddocuments the public tools exposed by production.use-cases/is generated from Calven's source and replaced whole on every sync. Do not hand-edit it.docs/assets/mcp-diagram.pnganddocs/assets/hero.pngare rendered from Calven's brand templates; re-render rather than edit. After re-rendering any PNG, runpython3 scripts/compress-images.py <files>: the plugin installs the whole repository, so image weight lands on every user's disk.https://api.calven.ai/mcpis the only production endpoint.
Keep mcp.json, .mcp.json, examples, documentation, and skills aligned with that endpoint. Run npm ci --prefix scripts once, then npm run validate --prefix scripts after every change.
Boundaries
- Never add a credential, tenant identifier, customer name, or private server source.
- Record a clean-account test in
docs/client-support.mdfor every listed client. - Do not add or rename a documented tool without verifying the production catalog.
- Skills must distinguish workspace evidence from model knowledge and must not invent missing evidence.
- Security reports belong in the private channel named in
SECURITY.md, not public issues.
Writing
Use the bare brand name Calven. Use calven.ai only when referring to the domain. Keep prose direct and specific. Avoid unsupported claims and promotional filler.
Terminology: "sign in" (not "authenticate"); "Client ID Metadata Document (CIMD)" on first use, then CIMD; "dynamic client registration (DCR)"; "battlecard".
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.