> ## Documentation Index
> Fetch the complete documentation index at: https://predev.mintlify.site/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: npx -y @predotdev/mcp setup adds it to every local coding agent with every pre.dev tool plus the user's own Chrome; the hosted server is https://api.pre.dev/mcp. Search these docs over MCP at https://docs.pre.dev/mcp.

# Fast vs Deep specs

> Choose planning depth based on the decisions you need to make.

Fast Spec gives a quicker project-level plan. Deep Spec develops a more detailed breakdown for complex work. Both can produce requirements, architecture, and specifications for humans and coding agents.

| Choose | When it helps |
| - | - |
| Fast Spec | Explore scope, validate an idea, or establish a first architecture |
| Deep Spec | Coordinate a larger project or examine implementation details and subtasks |

Deep output typically includes more granular stories and subtasks. Inspect the generated artifacts for the actual level of detail; do not rely on a fixed number of milestones or tasks.

Generation time and credits vary with scope. The [Architect overview](https://predev.mintlify.site/architect-agent/overview.md) keeps current estimates in one place. [Plans and credits](https://predev.mintlify.site/coding-agent/plans-and-credits.md) explains commercial availability.

## Refine the plan

Start with Fast when you need to assess direction, then request deeper planning for areas that need more detail. In the REST API, an existing spec's text goes in `currentContext`; in MCP it goes in `existingContext`. Neither field is a spec ID or an automatic upgrade operation.

In the Coding Agent, planning happens when the agent decides a request needs it, and you can ask for a plan at any time. That is separate from spec depth and [model choice](https://predev.mintlify.site/coding-agent/building/pro-mode.md).

See [understanding specs](https://predev.mintlify.site/coding-agent/specifications/understanding-specs.md) and [API inputs and outputs](https://predev.mintlify.site/architect-agent/inputs-and-outputs.md).


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