> ## 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.

# Understanding specs

> Read requirements, architecture, milestones, and the artifacts available to agents.

A spec connects the project's goals to a proposed implementation. It supplies context for planning, coding, and review; it should be checked against the actual requirements and current codebase.

A useful reading order is **goal → architecture → milestone → story → acceptance criteria**. Deep specifications can add implementation subtasks to a story.

An illustrative outline, with the subtasks a Deep spec adds:

```text theme={null}
Goal                  Keep a small team's work visible
Architecture          Web app, API, and project data
Milestone             A working issue board
  Story               Filter issues by assignee
    Acceptance        Only matching issues appear; an empty view explains why
    Subtasks (Deep)   Add the filter control; apply the assignee predicate;
                      check the matching and empty states
```

## Read the artifacts

| Artifact | What to review |
| - | - |
| Executive summary | Intended users, scope, and constraints |
| Core functionality | The behaviors the project should support |
| Architecture and tech stack | Components, dependencies, and technology choices |
| Milestones and stories | Delivery order and concrete units of work |
| Acceptance criteria | Observable outcomes used in verification |
| Subtasks | Implementation detail, especially in Deep specs |

Human-oriented output can include roles, personas, and effort estimates. Coding-agent output emphasizes implementation and omits human staffing detail. The API exposes Markdown, structured JSON, and available graphs; see [inputs and outputs](https://predev.mintlify.site/architect-agent/inputs-and-outputs.md) for exact fields.

## Refine before building

Ask the agent to resolve missing requirements, adjust constraints, or refine acceptance criteria. Use the available plan and story editing controls for the project. Confirm the updated artifact before launching work that depends on it.

Changing a plan does not retroactively update shipped code. Request the corresponding implementation change, and check that an active task is using the intended requirements.

## Example acceptance criteria

```text theme={null}
Story: Export invoices
- The exported rows match the current filters.
- The file includes the currency for every monetary amount.
- Users cannot export invoices outside their account.
- An empty result displays an explanation instead of an empty download.
```

Track planned work in the [roadmap](https://predev.mintlify.site/coding-agent/building/progress-tracking.md) and inspect implementation evidence in [verification](https://predev.mintlify.site/coding-agent/building/acceptance-criteria.md).


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