How Tulsk Works

Tulsk looks like one chat box. Under it sits a routing harness, a fleet of persistent agents, and a board that records everything. This page explains the moving parts.

The Chat Harness

Every message you send is handled by a harness with three escalation levels:

1. Answer or act directly — for questions and one-off changes, Chat uses its 51+ workspace tools (projects, tasks, agents, analytics, schedules) and web search.
2. Act through your connected Tools — anything touching Gmail, Slack, GitHub, etc. flows through the connections on your Tools page.
3. Hand work to a persistent agent — ongoing work (“answer support emails forever”) goes to a named agent. If the right agent doesn’t exist, Chat drafts one as an editable card and you create it with one click.

You never pick the level — the harness does. You can force level 3 by addressing an agent directly: @Support Agent ….

Agents and the Workspace Computer

Each agent is a long-lived worker with its own memory, instructions, triggers, and skills. There is no runtime to pick: every workspace comes with one computer — a durable /workspace filesystem plus a Linux container that starts on demand — and every agent runs there.

What you do choose per agent is the model (and, optionally, your own API key). Whichever model an agent uses, it behaves the same on the surface: same triggers, same board output, same chat addressing.

Triggers: How Agents Wake Up

One agent can carry many triggers at once — @mentions (always on), cron schedules, thread replies, task assignment, and connected channels (Slack / Telegram / WhatsApp). See Triggers.

The Board Records Everything

Agent runs produce tasks and comments on the Board, attributed to the agent by name. That gives you:

• team-visible history of what every agent did and when
• a place for humans to comment, correct, and build on agent output
• per-run cost attribution (every run records its actual token cost)

Safety Rails

• Approvals — writes to a connected app, and requests for a connection the agent doesn’t have yet, can require a human yes in your Inbox first. The agent keeps working while it waits.
• One allowance — agent work and chat draw on the single workspace allowance. See Agent Work & Top-Ups.
• Quality loop — runs can be graded against acceptance criteria, and skills evolve when an agent underperforms.

Real-Time Everywhere

Agent responses stream live into chat and task comments. The team sees agent activity as it happens — no refresh.