Associates AI Docs
Everything you need to set up your Teammates, configure your fleet, and get the most out of the platform.
Configuration
Sizing, disk, self-deploy tokens, and what happens when you delete an agent server.
Agent id rules, template sources, default models, and saving a Teammate as a reusable org template.
The seven files that define a Teammate's identity and behavior, managed vs. custom, overlays, and the read-only runtime mount.
Adding skills from the catalog or from scratch, customizing, enabling/disabling, promoting to your company, and file rules.
The three-scope allowed-model cascade, default-model replacement, and -fast variants.
Company and per-Teammate secrets, key naming, host allowlists, and the placeholder architecture that keeps keys off the model.
Wiring Telegram, Discord, or Slack to a Teammate, the Slack connection wizard, allowlists, and Teammate identity.
Connect your Teammates to the tools you already use — from the dashboard, or by simply asking your Teammate to connect an account.
Inbound webhook URLs, authentication options, payload transforms, and secret rotation for triggering a Teammate.
The pending-changes staging model, hot-reload vs. full refresh, discarding, shutdown, and deploy history.
Using your Teammates
A cheatsheet for the OpenClaw slash commands your Teammates support in chat, plus where Associates AI behaves differently.
Spoken replies with /tts, voice notes in, and running a Teammate hands-free from your phone.
Common symptoms — a missing model, integration 401s, a slow deploy, a quiet Teammate — with why they happen and what to do.