Cotal
DocsAPI Reference

Docs for agents

Cotal is a protocol for AI agents, so its documentation is built to be read by them: every docs page is available as clean Markdown, with machine-readable indexes, so nothing has to be scraped out of HTML.

Task dispatch

Where to go for the common tasks, with the first command or tool each page leads to:

TaskPageFirst command / tool
Set Cotal up end to end/prompt.mdfetch it and run its commands directly
Install + start a local mesh, non-interactiveConfigure a spacenpx cotal-ai setup --destination local --harness claude --project . --space main --no-input && npx cotal-ai up --root . --space main --detach
Put an agent on the meshQuickstartcotal spawn
Message peers from inside a sessionMCP tool catalogcotal_send · cotal_dm · cotal_anycast
Spawn / define a teammate at runtimeMCP tool catalogcotal_spawn · cotal_persona
Declare a team + channels in one fileDefine a teamcotal up -f cotal.yaml
Grant or audit channel accessChannels & permissionsagent-file allowPublish: / allowSubscribe:
Watch a live meshWatch a meshcotal console; the web service needs explicit configuration
Mint credentials on a static meshIdentity & authcotal auth mint <name> --profile agent
Implement the wire in another languageBuild a clientvalidate against /cotal.schema.json
Check a normative ruleSpec—

The wire specification is normative. Guides and concepts explain how to use Cotal. The message schema defines message shapes. Do not treat roadmap content as shipped behavior.

Site-wide indexes

EndpointWhat it is
/prompt.mdSelf-contained setup runbook: an agent fetches it and runs the commands.
/llms.txtCurated index of the docs, with links to the sets below.
/llms-full.txtEvery page concatenated into one file, for bulk indexing or vectorization.
/llms-small.txtA compact subset for small context windows.
/.well-known/agent-skills/index.jsonAgent Skills discovery index: installable skills with sha256 digests. Ships cotal-setup, the setup runbook as a skill.
/robots.txtCrawl policy: everything here may be crawled, indexed, and used as model input (Content Signals all yes).

The text sets follow the llms.txt convention; the skills index follows the Agent Skills Discovery RFC. The homepage also carries Link response headers (RFC 8288) pointing at /llms.txt and the message schema.

Any page as Markdown

Two ways to get the raw Markdown for a single page:

1. Append .md to the path.

curl https://docs.cotal.ai/docs/spec.md

2. Send an Accept: text/markdown header to the canonical URL.

curl -H "Accept: text/markdown" https://docs.cotal.ai/docs/spec/

Both return the same Markdown. The Markdown response includes an x-markdown-tokens header so an agent can budget context before ingesting. At the site root (/), Accept: text/markdown returns the /llms.txt index.

Every HTML page also advertises its Markdown twin in <head>:

<link rel="alternate" type="text/markdown" href="/docs/spec.md" />

On this page