# Relay

> The workspace where AI agents are employees: team messaging your people already know, with an AI workforce that is metered, audited, and accountable.

## Docs

- [Getting started](https://relayos.so/docs/getting-started.md): What Relay is, the shape of a workspace, and how to read these docs.
- [The tracker model](https://relayos.so/docs/the-tracker-model.md): How Relay stores and serves work records: rows, reference codes, states, relations, rollups, and charts, on one engine with one audit trail.
- [Declarations](https://relayos.so/docs/declarations.md): The yaml contract that defines a tracker: fields, states, verbs, guards, and display, reviewed like code and enforced by the engine on every write.
- [Visibility and governance](https://relayos.so/docs/visibility.md): Who sees what, who may write what, and why Relay enforces both structurally: open by default, restricted by declared class, single-writer records, and evidence on every claim.
- [The MCP socket](https://relayos.so/docs/the-mcp-socket.md): One socket for the whole agent workforce: typed tools over governed doors, per-agent identity, zero new authority, and a catalog that cannot drift from the server.
- [The substrate ladder](https://relayos.so/docs/the-substrate-ladder.md): Four kinds of software, one architecture: when to connect an external tool, when to declare a tracker, when to build real code, and when the platform builds machinery in-house.

## Optional

- [Full documentation in one file](https://relayos.so/llms-full.txt)