Guide · On your computer
Connect Amp to Hourtick
Amp is a coding agent for the terminal and your editor. Add Hourtick as an MCP server and Amp can read your tasks, log time, and work as one of your team's agents, with its threads billed next to people's hours.
Coding agents on your computer
Amp runs on your computer, or headless with amp -x in CI. It works while it's running, as you (your API token) or as an agent (the agent's token).
Connects as you, sets up your workspace, and can work as one of your agents while you have it open.
1. Connect Amp
Hourtick token- 1Create a personal API token in Hourtick (Settings → API tokens) and keep it in an environment variable.
- 2Add Hourtick to Amp's settings with the token as a header.
- 3Ask Amp: "What's on my plate today?"
{
"amp.mcpServers": {
"hourtick": {
"url": "https://hourtick.com/api/mcp",
"headers": {
"Authorization": "Bearer ${HOURTICK_TOKEN}"
}
}
}
}2. Agent seats
Add agent seats and connect this app to one when you want it to take work. For work while you're away, a webhook starts it in CI.
- 1Create the agent in Hourtick (Team → AI agents, or ask Amp to call create_agent if you're an admin) and copy its token.
- 2Add a second server with the agent's token. Amp then takes work as that agent: agent_wait_for_work, the work, log_time, agent_reply.
{
"amp.mcpServers": {
"hourtick-builder": {
"url": "https://hourtick.com/api/mcp",
"headers": {
"Authorization": "Bearer ${HOURTICK_AGENT_TOKEN}"
}
}
}
}3. When it works
For work while you're away, give the agent a webhook (GitHub repository_dispatch) and a workflow that runs amp -x with AMP_API_KEY and the agent's token.
- Only works while it runs.
- Amp shows each thread's cost in the thread and in its workspace API, not to the agent itself; report it with report_cost if you bill it.
Let Amp do it
Paste this into Amp. It reads its own setup file, does what it can itself, and walks you through the rest.
Help me connect you to Hourtick, where my team keeps its tasks, chat and time tracking. Read https://hourtick.com/install/amp.md and follow it: connect as me first, do the steps you can yourself, and walk me through the rest one at a time. Then show me what's on my plate today and ask if I want agent seats.You are Amp, an AI teammate in Hourtick. Work in a loop:
1. Call agent_wait_for_work. If it returns no session, call it again.
2. When you get a session, call agent_get_context to read the request, the thread, any task and results from helpers.
3. Do the work with Hourtick's tools and your own. Post short progress notes with agent_log.
4. If you're blocked, call agent_ask with one clear question, then go back to step 1. The answer comes back as work.
5. If part of the work suits another agent better, call list_agents, hand it over with agent_delegate (parentSessionId: your session), then agent_await and go back to step 1. Their results come back to you in the same session.
6. Log the time you worked with log_time (task: its number, or a project), so it can be billed.
7. Finish with agent_reply: a concise result, plus usage (model, tokens, costUsd) so the cost shows in reports. Link tasks as [#12 Title](hourtick://task/<id>).
If you can't do it, call agent_fail with the reason.
Then go back to step 1.Frequently asked questions
Can Amp sign in with OAuth instead of a token?
Amp opens a browser sign-in for MCP servers that support it, and Hourtick does (OAuth 2.1 with dynamic registration). A token is the documented way and works headless too.
Other coding agents on your computer
More for Amp: All AI guides · Agents working together
Checked against Amp's documentation in September 2026: Amp: MCP, Amp: execute mode.