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

# Octopoda: security scanning for apps built with AI

> Octopoda scans your code for security holes, shows the evidence for each one, and gives you a fix prompt to paste into Cursor or Claude Code.

Octopoda finds the security holes in your app before someone else does. It reads your code, runs it in a private sandbox, and reports only the issues it could reproduce, plus dependencies with a published vulnerability. Each finding comes with its evidence and a fix prompt you can paste into Cursor or Claude Code.

If you build fast with AI coding tools, Octopoda is the security review you don't have time to do by hand: scan before you ship, and let pull request checks catch new problems before they merge.

## Ways to use Octopoda

> [card] [Command line](https://docs.octopoda.dev/scanning/code-scans.md)
>
> Run `octopoda scan` in any project folder and follow the scan from your terminal.

> [card] [Pull request checks](https://docs.octopoda.dev/scanning/pull-request-checks.md)
>
> Scan every pull request on GitHub, with a check and one summary comment.

> [card] [Your AI coding tool](https://docs.octopoda.dev/integrations/mcp.md)
>
> Let Cursor or Claude Code start scans and read fixes through the MCP server.

> [card] [Web app](https://docs.octopoda.dev/findings/read-a-finding.md)
>
> Review findings, proof and fix prompts with your team at app.octopoda.dev.

## What makes Octopoda different

* **Findings come with proof.** Octopoda reproduces an issue in your code before it reports it, so you don't spend your time on warnings that turn out to be nothing. See [How Octopoda works](https://docs.octopoda.dev/how-it-works.md).
* **Fixes are ready for your AI tool.** Each finding has a prompt that names the file and the change, written for Cursor or Claude Code. See [Fix a finding with your AI tool](https://docs.octopoda.dev/findings/fix-with-ai.md).
* **Your code stays private.** Each scan runs in its own isolated sandbox that is destroyed when the scan ends, and your code is never used to train AI models. See [How Octopoda handles your code](https://docs.octopoda.dev/security/your-code.md).

## Get started

> [card] [Quickstart](https://docs.octopoda.dev/quickstart.md)
>
> Start a trial, install the CLI and run your first scan.

> [card] [Plans and credits](https://docs.octopoda.dev/billing/plans-and-credits.md)
>
> See what each plan includes and what each scan costs.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.