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

# Agent Integration

> Call ADHD at decision points inside your own agent loop — the shape that pays the most.

The highest-leverage way to use the library: call `run()` at **decision points** inside a larger planning, coding, or review agent, and feed the deepened sketches back into the agent's context.

```ts theme={null}
import { run } from "adhd-agent";

// inside your planning / coding / review agent
if (agentIsAtADecisionPoint) {
  const { shortlist, nonObviousPick, traps, deepened } = await run({
    problem: framedDecision,
    context: relevantCode,
    framesPerRun: 4,
    topK: 2,
    codeMode: true,
  });
  // feed the deepened sketches back into your agent's context
}
```

## Good moments to call it

* An agent **stuck after N attempts** on a bug — widen the hypothesis space with new hypothesis *classes*
* A planning agent at a **branch point with high uncertainty**
* A code-review agent asked *"what could go wrong here"*
* A refactor agent picking **which abstraction to introduce**
* A test-generation agent generating **adversarial inputs** (the inversion frame is built for this)

## Programmatic skill injection (Agent SDK)

You can also inject the skill itself into a Claude Agent SDK session, letting the agent run the loop with its own Task tool instead of via the library:

```ts theme={null}
import { query } from "@anthropic-ai/claude-agent-sdk";
import { readFileSync } from "node:fs";

const skill = readFileSync("./skills/adhd/SKILL.md", "utf8");

for await (const m of query({
  prompt: "design a retry strategy for a CLI whose LLM hangs for 90s",
  options: {
    systemPrompt: { type: "preset", preset: "claude_code", append: skill },
    allowedTools: ["Task"],
  },
})) {
  // …
}
```

## Which shape to pick

<CardGroup cols={2}>
  <Card title="Library run() call" icon="code">
    Minimal token substrate — each branch carries only the problem + frame prompt. Structured `RunResult` back. Best for batch and embedding in pipelines.
  </Card>

  <Card title="Skill inside a session" icon="wand-magic-sparkles">
    The agent orchestrates the frames itself with fresh-context Task calls. Zero extra dependencies, but every branch re-loads the session's base context — see [the cost model](/concepts/when-to-use#cost--speed).
  </Card>
</CardGroup>

## Real-world integrations

Projects have wired ADHD into their own loops in several shapes — mesh-orchestrator peers ([repowire](https://github.com/prassanna-ravishankar/repowire)), a marketplace `think` plugin ([mstack](https://github.com/mayank-io/mstack)), an anchoring-bias pre-pass before review ([zk-flow-oss](https://github.com/matt-metivier/zk-flow-oss)), and a codebase-onboarding explanation engine ([wtfismyrepo](https://github.com/nandnijaiswal/wtfismyrepo)). See [Adopters](/community/adopters) for the full list.