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

# JD Codec

> Compress browser-agent perception streams ~80%+ via a stateful, lossy cloud codec — with on-device PII redaction.

JD Codec sits between your AI agent and the browser it's automating. It compresses each DOM/ARIA snapshot the agent sees, sending the smallest possible representation while keeping every reference your agent needs to act. Token cost drops, accuracy holds.

## How it works in 30 seconds

1. **Install** the local connector (`npm install -g jdcodec` or `pip install jdcodec`).
2. **Wire it** to your AI agent as an MCP server (`claude mcp add --scope user jdcodec -- jdcodec`).
3. **Run your agent normally.** The connector intercepts browser snapshots, redacts PII on-device, sends the rest to the JD Codec cloud, splices the compressed output into the agent's tool response.

Your agent sees only what changed since the last step. Your customers' PII never leaves the local machine.

## Where to next

<CardGroup cols={2}>
  <Card title="Quickstart" icon="rocket" href="/docs/quickstart">
    Install, configure, and see your first compression event in under 5 minutes.
  </Card>

  <Card title="API Reference" icon="code" href="/docs/api-reference">
    Endpoints, schemas, error codes — everything the connector talks to.
  </Card>

  <Card title="What is JD Codec?" icon="circle-question" href="/docs/concepts/what-is-jdcodec">
    The codec, the connector, the cloud — what each piece does.
  </Card>

  <Card title="Privacy posture" icon="shield" href="/docs/concepts/privacy-posture">
    What stays local, what reaches the cloud, what's never persisted.
  </Card>
</CardGroup>