> ## Documentation Index
> Fetch the complete documentation index at: https://docs.m9tz.de/llms.txt
> Use this file to discover all available pages before exploring further.

# Handoff Phase: Own and Operate Your Agent After Launch

> Learn what you receive at handoff, how team training is structured, what ongoing support options are available, and how to request changes after launch.

A successful project doesn't end with a working agent — it ends with your team confident and capable of running it without us. The Handoff phase is where that independence is built. You receive full documentation, a recorded walkthrough, and a live training session so that owning the agent feels straightforward from day one.

## What You Receive at Handoff

At the close of the project, we hand over three core deliverables. These are yours to keep, share internally, and reference whenever you need them.

<CardGroup cols={3}>
  <Card title="Agent Runbook" icon="book-open">
    A plain-language document covering how the agent works, what it does in each scenario, how to read its logs, and how to interpret the most common outputs and alerts.
  </Card>

  <Card title="Admin Guide" icon="sliders">
    Step-by-step instructions for the tasks your team will actually need to do: updating the agent's responses, adjusting escalation thresholds, and pausing or resuming the agent.
  </Card>

  <Card title="Recorded Walkthrough Video" icon="circle-play">
    A screen-recorded video of us walking through the agent's configuration, key settings, and day-to-day operation — so your team can revisit it any time without scheduling a call.
  </Card>
</CardGroup>

<Note>
  All three deliverables are provided in editable formats where possible. You can update the Runbook and Admin Guide as your workflows evolve — they're living documents, not locked PDFs.
</Note>

## Team Training

Within the handoff period, we run a **60-minute live training session** for up to five of your team members.

<Steps>
  <Step title="Day-to-day use">
    We walk through what your team will actually see and interact with during normal operations — how the agent surfaces its work, where to check on it, and what healthy activity looks like.
  </Step>

  <Step title="Monitoring and log reading">
    We show your team how to check the agent's activity logs, how to spot errors or anomalies, and when something warrants a closer look versus when it's expected behaviour.
  </Step>

  <Step title="Escalation handling">
    We cover the escalation paths built into the agent — what triggers a handoff to a human, how that handoff is communicated, and how your team should respond when one comes in.
  </Step>

  <Step title="Live Q&A">
    The final portion of the session is open for questions. Bring your most sceptical team member — the more questions asked here, the more confident everyone walks away.
  </Step>
</Steps>

<Tip>
  Record the training session for team members who can't attend live, or for future onboarding when new staff join the team. Let us know before the session and we'll make sure the recording is clean and shareable.
</Tip>

## Ongoing Support Options

After handoff, you have two paths forward depending on how much involvement you want from us.

<Tabs>
  <Tab title="Ongoing Support Arrangement">
    We offer an optional ongoing support arrangement, scoped and agreed individually, that keeps us actively involved in your agent's health. This includes regular monitoring, performance reviews, minor updates as your workflows evolve, and priority response if something needs attention. It's a good fit if your workflows change frequently or if the agent is business-critical.
  </Tab>

  <Tab title="Self-Managed">
    If you'd rather run independently, you're fully equipped to do so. The Runbook, Admin Guide, and training session give you everything you need to monitor and maintain the agent on your own. You can always re-engage us for a specific request without committing to an ongoing arrangement.
  </Tab>
</Tabs>

## How to Request Changes After Handoff

Workflows evolve, and your agent should be able to evolve with them. After handoff, submit a **change request form** (provided at handoff) to kick off any update.

<Accordion title="How change requests are scoped">
  **Small changes** — wording updates, threshold adjustments, minor response tweaks — are changes within the existing logic of the agent that don't require new integrations or workflow branches. These are typically straightforward to handle.

  **Workflow additions** — new triggers, new tool integrations, new decision branches, or new escalation paths — are scoped as a separate engagement. We'll review your request, provide a timeline and effort estimate, and treat it as a mini design-and-build cycle before implementing anything.

  In both cases, we'll respond to your change request within two business days with a clear indication of which category it falls into and what the next step is.
</Accordion>

<Note>
  Not sure whether your change is "small" or "a workflow addition"? Submit the form and describe what you want to change in plain language. We'll categorise it for you.
</Note>

## Data Ownership

Everything processed by your agent belongs to you — always.

<CardGroup cols={2}>
  <Card title="Your Data, Your Control" icon="lock">
    All inputs, outputs, and logs generated by the agent are stored in your environment or in storage you own. We don't retain your business data after the project closes.
  </Card>

  <Card title="Export and Migration Tools" icon="file-export">
    If you ever want to migrate to a different platform or take the agent in-house, we provide export tools and full technical documentation to make that transition straightforward.
  </Card>
</CardGroup>

<Warning>
  Before the project closes, confirm with your team where agent logs are stored and who has access. This is covered in the Admin Guide, but it's worth an internal conversation so the right people have visibility from the start.
</Warning>

You've completed all five phases: [Discovery](/blueprint/discovery), [Design](/blueprint/design), [Build](/blueprint/build), [Deploy](/blueprint/deploy), and Handoff. Your agent is live, your team is trained, and your documentation is in hand.