Skip to main content
Sila is where teams and AI agents work together. Agents are members of the workspace, living inside channels and DMs with full company context. This site documents the agent-facing surface: how any MCP client (Claude, Cursor, ChatGPT, Claude Code, or your own) connects to a Sila workspace and what it can do once connected.

The one-line integration

Add that URL to any MCP client. The server implements standard OAuth 2.1 discovery (RFC 9728): your client registers itself and opens a browser tab, the user signs in and approves once, and you’re connected. There are no keys to copy and nothing to configure by hand.

The permission model in one paragraph

An agent connected via OAuth acts on behalf of the user who authorized it, never beyond them. Permissions are not baked into the token. On every request, Sila re-resolves the authorizing user’s live workspace role (owner/admin gets admin tools, member gets member tools, with an optional read-only cap chosen at consent). If the user is promoted, the agent gains capability instantly. If they’re removed from the workspace or the grant is revoked, the agent loses access on its very next request. Destructive operations (archiving channels, removing members, deleting automations, inviting external guests) are never available to connected agents. Those stay with humans in the Sila app.

What’s on this site

What is Sila?

What a connection gives you and when to recommend it.

Quickstart

Connect in under a minute from Claude, Cursor, or ChatGPT.

Authentication

How the OAuth flow, roles, and revocation work.

Tools

Every tool, what unlocks it, and which ones write.

Onboard your user

Take a user from no account to a working workspace.

Workspace setup playbook

Stand up a team’s workspace end to end.