Skip to main content

​
Modes

Headmaster has four interaction modes that control how much autonomy the agent has and how often it asks for your approval. Switch modes from the mode selector in the composer area. You can also switch mid-conversation — the change applies to the next action, not retroactively.

​
Work Along

The default mode. The agent works alongside you, asking for approval before running tools, writing files, or making external calls. You see every action before it happens. Use when:
  • 🔍 Exploring a new codebase and wanting to see what the agent does.
  • 🔒 Working on something sensitive (production config, customer data).
  • 🎓 Learning from the agent’s approach.
How it works:
  • The agent proposes an action, shows its parameters and expected outcome.
  • You click Approve to let it run, or Deny to stop it.
  • Use Approve always to greenlight all future calls of the same type for this conversation.

​
Run It Yourself

The agent handles the task end-to-end without asking for approval on each step. It still asks before doing anything irreversible. Use when:
  • ✅ You’ve given a clear task and trust the agent to execute.
  • ⚡ The task has many steps and approving each one would be tedious.
  • 🔁 Running a well-understood workflow.
How it works:
  • The agent runs tools and writes files without per-action approval.
  • It still asks before destructive actions (delete, overwrite, send, push).
  • You can interrupt at any time with the stop button.
  • All actions are logged in the conversation transcript for audit.

​
Hands-Off

Same as Run It Yourself, but the agent does not ask before destructive actions either. It just does them and reports what it did. Use when:
  • 🤖 Running a well-tested automated workflow.
  • 🚶 You’re away from the keyboard.
  • 🛡️ You’ve set up guardrails (project folder scoping, allowed tools list) and trust the setup.
How it works:
  • No approval prompts at all. The agent runs to completion.
  • Every action is still logged in the transcript.
  • If the agent encounters an error it can’t resolve, it stops and reports rather than guessing.
Hands-Off mode will delete files and run commands without asking. Set it up on well-understood workflows only. Test in Run It Yourself first.

​
Autopilot

The agent runs on a schedule, 24/7. It picks up tasks from the Schedule, executes them in Hands-Off mode, and delivers results to your configured Channels. Use when:
  • ⏰ You have recurring tasks (daily summaries, weekly reports, monitoring).
  • 🌙 You want the agent to work while you sleep.
  • ✅ You’ve validated the workflow in Hands-Off first.
How it works:
  1. The agent wakes on a schedule trigger.
  2. It runs the task in Hands-Off mode (no approvals).
  3. Results are delivered to the channel you configured (Telegram, email, desktop notification).
  4. The agent goes back to sleep until the next trigger.
Always test a workflow in Hands-Off mode manually before scheduling it in Autopilot. See The Schedule for setup.

​
Approval gates

In Work Along and Run It Yourself, the agent asks for approval before:
  • 🔧 Running a tool — web search, code execution, file read/write, MCP call.
  • 📁 Accessing a file outside the project folder.
  • 🌐 Making a network request (external API call, HTTP request).
  • ⚠️ Destructive actions (delete, overwrite, send, push) — asked in all modes except Hands-Off.
The approval prompt shows:
  • What the agent wants to do.
  • The tool name and parameters.
  • The expected outcome.
  • Approve / Deny / Approve always (for this conversation) buttons.

Approvals

Full details on the approval flow — clarifications, permissions, secret requests.

The Schedule

Set up Autopilot with cron-based recurring tasks.