> For the complete documentation index, see [llms.txt](https://docs.zerve.ai/zerve-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/evolve-mode/specialists.md).

# Specialists

The specialist agents Evolve Mode delegates work to.

Evolve Mode hands parts of a task to **specialists**: agents that each focus on one kind of work. Several can run at once, and each reports back to Evolve Mode, which checks their work.

| Specialist            | What it does                                                                                                                                        |
| --------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Coding**            | Does the work in the notebook: writing and running code, fetching data, building models. Several can run in parallel on different parts of a task.  |
| **Environment setup** | Installs packages into the notebook's environment and waits for the build to finish                                                                 |
| **Research**          | Searches the web and academic literature, finds datasets and builds a library of citable references                                                 |
| **Ideation**          | Brainstorms approaches, including ideas borrowed from other fields such as physics, biology or control theory, and recommends the best two or three |
| **Peer review**       | Reviews work without changing it, and either approves it or asks for changes                                                                        |
| **Drafting**          | Writes documents in Markdown, with citations. Each revision is saved as a new version of the same document.                                         |
| **Custom subagents**  | Specialists your organization has defined for its own work. See [Custom subagents](/zerve-docs/agents-in-the-workspace/custom-subagents.md).        |

## Research

The research specialist searches the web and academic sources, including OpenAlex, Crossref, PubMed, Europe PMC, bioRxiv and medRxiv, arXiv, Semantic Scholar, Zenodo and DataCite. The references it finds are saved to the notebook's [Sources](/zerve-docs/agents-in-the-workspace/sources-and-citations.md), so the agent can cite them.

{% hint style="info" %}
The research specialist is only available when your organization uses Anthropic or Amazon Bedrock as its model provider. See [Supported models](/zerve-docs/agents-in-the-workspace/supported-models.md).
{% endhint %}

## Drafting

The drafting specialist writes reports, memos and other documents as Markdown, with citations to the sources the agent used. Documents appear as file outputs in the notebook, and each revision adds a new version rather than replacing the old one.

For Word or PowerPoint files, Evolve Mode uses a coding specialist instead, which builds the file in code.

## Following specialists

Each specialist appears in Evolve Mode's plan with its status:

| Status        | Meaning                    |
| ------------- | -------------------------- |
| **Running**   | Working now                |
| **Completed** | Finished                   |
| **Failed**    | Couldn't complete its work |
| **Cancelled** | Stopped before finishing   |


---

# 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.zerve.ai/zerve-docs/agents-in-the-workspace/evolve-mode/specialists.md?ask=<question>&goal=<user_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.