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

> ## Agent Instructions
> Agents: start with https://docs.pre.dev/agents.md, which has complete recipes, plan access, polling rules, errors and limits.
> Authenticate with the workspace API key (pdk_…) from Integrations → Built-in, sent as Authorization: Bearer <key>.
> REST API: https://api.pre.dev (OpenAPI: https://docs.pre.dev/api-reference/openapi.json). AI Gateway: https://api.pre.dev/v1, OpenAI-compatible (OpenAPI: https://docs.pre.dev/api-reference/ai-gateway.openapi.json).
> pre.dev MCP server: https://api.pre.dev/mcp. Search these docs over MCP at https://docs.pre.dev/mcp.

# Architect API

> Generate software specifications with Markdown, structured JSON, architecture graphs, and implementation estimates.

The Architect API turns a project description into an implementation plan. Choose **Fast Spec** for milestones and stories, or **Deep Spec** for a more detailed breakdown with subtasks. Both accept existing codebase context and reference documentation.

<CardGroup cols={2}>
  <Card title="Generate your first spec" icon="play" href="/architect-agent/quickstart">
    Submit a request, poll its status, and read the completed artifacts.
  </Card>

  <Card title="Use it from an assistant" icon="plug" href="/architect-agent/mcp-setup">
    Connect fast\_spec and deep\_spec through the pre.dev MCP server.
  </Card>
</CardGroup>

## From request to artifacts

Send your requirements, optional codebase context, and any reference files or documentation URLs. The completed result offers the same plan in formats suited to different readers.

The coding-agent variant focuses on requirements and implementation structure. The human variant adds personas, roles, and effort estimates. Graphs, an architecture infographic, and documentation archives are returned when available; integrations should tolerate absent optional artifacts.

## Choose the depth

| | Fast Spec | Deep Spec |
| - | - | - |
| Structure | Milestones and user stories | Milestones, stories, and subtasks |
| Starting point | Early planning and straightforward features | Detailed implementation planning |
| Typical generation time | About 1 minute | About 3–5 minutes |
| Typical credits | About 5–10 | About 10–50 |

Times and credit usage are estimates, not fixed limits or quotes. Use the returned `creditsUsed` for actual consumption. See [plans and credits](/coding-agent/plans-and-credits) for access, and [Fast vs Deep](/coding-agent/specifications/fast-vs-deep) for planning guidance.

## Read next

* [Inputs and outputs](/architect-agent/inputs-and-outputs): files, context, documentation, and graph data.
* [Status and results](/architect-agent/api/spec-status): polling and failure handling.
* [Project plan](/architect-agent/api/get-plan): what is built, what works, what is left, and where it lives in the code.
* [List](/architect-agent/api/list-specs) or [search](/architect-agent/api/find-specs): retrieve existing work.
* [Python](/architect-agent/sdks/python) and [Node](/architect-agent/sdks/node): official SDKs.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.