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

# deeda Think

> A desktop knowledge and agent platform with a state-of-the-art multi-provider agent harness.

**deeda Think** is a desktop knowledge/agent platform. Its agent harness drives
foundational models from **Anthropic, OpenAI, Google Gemini, and local
runtimes** to the full extent of their capabilities — all governed from a
single declarative control plane: **workflow markdown**, validated and
dispatched by **Cadence**.

<CardGroup cols={2}>
  <Card title="Harness Overview" icon="diagram-project" href="/harness/overview">
    How the pieces fit: Cadence, workflow markdown, runtimes, broker, compiler.
  </Card>

  <Card title="Quickstart" icon="bolt" href="/quickstart">
    Author and dispatch your first cross-provider workflow.
  </Card>

  <Card title="Workflow Schema" icon="list-tree" href="/harness/workflow-schema">
    Every field Cadence validates — types, enums, defaults. The `--help` of workflows.
  </Card>

  <Card title="Cookbook & Snippets" icon="puzzle-piece" href="/harness/cookbook">
    Complete recipes plus composable blocks to assemble your own workflow.
  </Card>
</CardGroup>

<Note>
  **AI agent?** Read [For AI Agents](/agents) — three fetches are enough to
  assemble and dispatch any workflow.
</Note>

## Design principles

The harness is built on a small set of non-negotiables: one source of truth
per concept, no speculative abstractions, behavior-preserving changes by
default, and **proof over assertion** — a capability counts as available only
when a real probe exercises it and the evidence is checked in. The docs you
are reading inherit that discipline: the data-heavy pages are generated
directly from the same catalogs the harness executes against, so they cannot
drift from reality.