---
title: "Code Mode"
description: "Let an agent write a short script against the Lookio API in a sandbox, with two tools, search and execute, the same scopes and strict limits."
canonical: "https://lookio.app/docs/developers/mcp/code-mode"
updated: "2026-11-08"
---

# Code Mode

Code Mode is a second MCP server, for agents that work better when they write code than when they chain many tool calls.

```text
https://mcp.lookio.app/mcp/code
```

It has two tools. The standard server at `/mcp` stays the default.

| Tool      | What it does                                                                                                                                                                                |
| --------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `search`  | Finds API operations. The agent sends a JavaScript filter that runs on the API description in the sandbox. Without a filter, it returns all operations in short form: a few hundred tokens. |
| `execute` | Runs a JavaScript program with a `request()` function that calls the API.                                                                                                                   |

The API description stays in the sandbox and does not fill the agent's context.

## Calling the API

In `execute`, `request({ method, path, query, body })` calls an endpoint and returns the parsed response: `{ data, request_id }`, and `has_more` and `next_cursor` for a list. A failed call throws an `Error` with the API's message.

```js title="Re-index every failed resource"
const failed = [];
let cursor;

do {
  const page = await request({
    method: "GET",
    path: "/v1/resources",
    query: { limit: 100, ...(cursor ? { cursor } : {}) },
  });
  for (const resource of page.data) {
    if (resource.status === "failed") {
      await request({ method: "POST", path: `/v1/resources/${resource.id}/reindex` });
      failed.push({ id: resource.id, title: resource.title });
    }
  }
  cursor = page.next_cursor;
} while (cursor);

return { reindexed: failed.length, failed };
```

Only what the script returns or logs with `console.log` goes back to the agent. Filter and summarise in the script. Do not return raw rows.

## Limits

| Limit                           | Value                                                              |
| ------------------------------- | ------------------------------------------------------------------ |
| Time for one execution          | 30 seconds                                                         |
| Size of a program               | 64 KB                                                              |
| Size of the result              | 256 KB. A longer result is cut and marked `[truncated by Lookio]`. |
| Log output                      | 200 lines and 32 KB                                                |
| `request()` calls per execution | 100. Going over fails the whole run.                               |

Every `request()` counts against the credential's [rate limit](/docs/developers/rate-limits), like the HTTP call.

## Same permissions, same costs

Code Mode uses the same routes, checks and scopes as an HTTP request. A path without its scope fails with `insufficient_scope`. A script cannot do more than the connection.

Costs are the same, and a loop multiplies them: 100 calls to `POST /v1/assistants/{assistant_id}/query` cost 100 answers. Do not grant `query` to a Code Mode connection without need, and tell the agent to ask before it loops over questions.

## What a script cannot do

- Open a network connection. There is no `fetch` and no file system. `request()` is the only way out.
- Upload a file or stream an answer.
- Set headers, including `Idempotency-Key`.
- Reach the engine, the database or another workspace.

## Connecting

Connect to the Code Mode URL like the standard server. See [Connect an AI client](/docs/developers/mcp/connect). A token for the standard server only does not work here, and the reverse. A token without `resource` works for both.

The sandbox protects against the code, not against the content. Text from your resources goes back to the agent and can try to change the next script. Narrow scopes limit the risk. See [Safety and permissions](/docs/developers/mcp/safety-and-permissions#content-is-untrusted).