> For the complete documentation index, see [llms.txt](https://docs.textmaddie.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.textmaddie.com/working-with-us/working-with-the-team.md).
# Working with the Text Maddie team
Text Maddie builds and maintains your iMessage agent. What to bring to the team, what you can do yourself, and how to report a bad reply so it gets fixed fast.
Text Maddie is a managed service. We build your agent, we maintain it, and we are the point of contact for anything that changes what the agent does.
You do not need to work out whether a request is reasonable or how it would be implemented. Describe what you want and we will tell you.
## What to bring to us
* Anything the agent should say, know, or stop saying
* A reminder cadence or follow-up schedule the agent should keep, described by what matters about the timing rather than by clock times. See [prompts.md](/concepts/prompts.md#times-in-prompts)
* A new agent, or a change to what an existing one is for
* Connecting a tool the agent should be able to use during a conversation
* Slack app setup
* A line that looks unhealthy. We monitor every line and usually act first, but tell us anyway
* A line you want retired: we move its conversations to healthy lines and release the number once they have moved
* Conversation styles that do not fit your brand
* Plans, pricing, and payment terms
* Turning features on or off, including prompt and style editing
* Anything that looks wrong and you cannot explain
## What you can do yourself
* Read conversations and send messages
* Pause and unpause the agent on a contact
* Set conversation behavior such as timezone and history depth
* Decide how a blocked line's conversations are rescued: automatically or by us, and whether leads who never replied hear from the new number
* Choose a conversation style, when style editing is on
* Manage Slack channels and your team directory
* Invite teammates and set roles
* Mint and revoke API keys
* Generate reports and export contact CSVs
## Reporting something wrong
The single most useful thing you can send is a **link to the run** from the [Activity](/your-dashboard/activity.md) page. It shows what the agent read, what it was thinking, and what it sent.
A good report:
```
Run:
The lead asked whether we service commercial properties.
The agent said no. We do, since March.
Happened twice this week.
```
That gives us the fix, its scope, and the evidence in three lines.
{% hint style="info" %}
For anything urgent, such as the agent saying something damaging, pause the contact first with **Pause AI**, then send the report. Pausing stops it immediately and you do not need to wait for us.
{% endhint %}
## Changes you can make safely right now
Some things are yours and do not need us:
* **A wrong note about one lead or customer.** Edit the scratchpad on that conversation.
* **A conversation that needs a person.** Pause the agent and take over.
* **A text at an odd hour.** Open the run from Activity. A reply to a lead or customer who was texting at the time is allowed at any hour; only the agent's own outreach waits for morning. If the agent started the exchange at night, send us that run.
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.
Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:
```
GET https://docs.textmaddie.com/working-with-us/working-with-the-team.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.