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

# AI Agents

> ChainStream's data and execution surfaces for AI agents — MCP Server, Agent Skills, CLI, and pay-per-call micropayments.

ChainStream exposes the same [data products](/en/docs/data-products/overview) and execution primitives to AI agents as to any other client — but with the authentication, tool discovery and pay-per-call patterns that agents actually need.

<Note>
  **AI Agents is a consumer surface, not a separate product.** Everything an
  agent can do is also available over REST, GraphQL, WebSocket, Kafka, the
  SDKs and the CLI. The pages in this section describe the shapes that make
  the catalog usable from LLMs.
</Note>

## Two main surfaces

<CardGroup cols={2}>
  <Card title="MCP Server" icon="robot" href="/en/docs/ai-agents/mcp-server/introduction">
    A hosted (and self-hostable) Model Context Protocol server that exposes
    ChainStream's catalog as typed tools. Works out-of-the-box with Claude,
    Cursor, ChatGPT, Windsurf and any MCP-compatible client.
  </Card>

  <Card title="Agent Skills" icon="brain" href="/en/docs/ai-agents/agent-skills/introduction">
    Installable SKILL.md packs that teach an agent <em>how</em> to use
    ChainStream — three skills cover standard data (`chainstream-data`),
    custom GraphQL analytics (`chainstream-graphql`) and DeFi execution
    (`chainstream-defi`). Ships ready for Cursor, Claude Code and Codex.
  </Card>
</CardGroup>

## Why this matters for agents

Traditional LLM access to blockchains fails in three ways: the model has a knowledge cutoff, it has no authenticated data source, and it cannot pay for reads without a pre-registered account. ChainStream solves each:

<AccordionGroup>
  <Accordion title="Live data, typed for tool use" icon="bolt">
    Every data product is reachable as a structured tool call with a JSON
    schema. The agent gets canonical IDs, not screen-scraped text — so it can
    chain calls without re-parsing on every step.
  </Accordion>

  <Accordion title="Authenticated access that fits agent loops" icon="shield-halved">
    API keys work, OAuth works, and for autonomous agents there is
    [x402 / MPP](/en/docs/platform/billing-payments/x402-payments) — a
    per-call HTTP payment scheme that means an agent can pay in USDC and
    call immediately, with no prior account setup.
  </Accordion>

  <Accordion title="Optional write paths with user-level guardrails" icon="right-left">
    DEX quotes, routing and swap submission are all available. Writes still
    require a signed transaction from a user-controlled wallet — agents can
    prepare transactions but not silently send from someone else's funds.
  </Accordion>
</AccordionGroup>

## Which surface should I pick?

| Situation                                                            | Surface                                                                          |
| :------------------------------------------------------------------- | :------------------------------------------------------------------------------- |
| You are using Claude / Cursor / ChatGPT and want ChainStream tools   | [MCP Server](/en/docs/ai-agents/mcp-server/introduction)                         |
| You want to install a ready-made agent skill for token research      | [chainstream-data skill](/en/docs/ai-agents/agent-skills/chainstream-data)       |
| You need cross-cube JOINs, custom aggregations or time-series        | [chainstream-graphql skill](/en/docs/ai-agents/agent-skills/chainstream-graphql) |
| You want the agent to execute swaps / create tokens                  | [chainstream-defi skill](/en/docs/ai-agents/agent-skills/chainstream-defi)       |
| You are writing your own agent in Python / TS and want direct access | [SDKs](/en/sdks/overview)                                                        |
| You want pay-per-call with USDC and no signup                        | [x402 Payments](/en/docs/platform/billing-payments/x402-payments)                |
| You want to pay with Tempo / Stripe / stablecoins                    | [MPP Payments](/en/docs/platform/billing-payments/mpp-payments)                  |
| You want shell-native tools the agent can orchestrate                | [CLI](/en/docs/access-methods/cli)                                               |

## Quickstart for agents

<Steps>
  <Step title="Start the MCP Server">
    ```bash theme={null}
    CHAINSTREAM_API_KEY=your_api_key npx @chainstream-io/mcp
    ```
  </Step>

  <Step title="Point your agent at it">
    Full configuration snippets for Claude Desktop, Cursor and ChatGPT are in
    [MCP Server Setup](/en/docs/ai-agents/mcp-server/setup).
  </Step>

  <Step title="Ask a question">
    *"Show me the top 10 Solana tokens by 24h volume and highlight any with
    KYT warnings."* The agent picks the right tools and chains the calls.
  </Step>
</Steps>

## Next

<CardGroup cols={2}>
  <Card title="MCP Server introduction" icon="robot" href="/en/docs/ai-agents/mcp-server/introduction">
    How the MCP Server is organized and what tools it exposes.
  </Card>

  <Card title="Agent Skills introduction" icon="brain" href="/en/docs/ai-agents/agent-skills/introduction">
    What Skills are, how they are installed and how to author new ones.
  </Card>

  <Card title="x402 Payments" icon="credit-card" href="/en/docs/platform/billing-payments/x402-payments">
    HTTP-native micropayments for agent workloads.
  </Card>

  <Card title="AI Agent tutorial" icon="graduation-cap" href="/en/docs/tutorials/ai-agent-with-mcp">
    Build an agent end-to-end using ChainStream.
  </Card>
</CardGroup>