# Code review with Cursor

URL: https://codepanel.ai/agents/cursor

Cursor, Anysphere's coding agent, in CodePanel's review loop: it connects through MCP and skills, reads Cursor rules and answers every finding.

Cursor is Anysphere's coding agent, in its own editor, a CLI, JetBrains IDEs through the Agent Client Protocol, cloud agents and an iOS app. With CodePanel it is the author on the other side of the review: after it pushes, it reads the panel's findings on that commit over MCP, fixes or disputes each one with evidence, and rates every finding. It reads .cursor/rules/*.mdc, AGENTS.md and CLAUDE.md, and the setup skill writes the loop into AGENTS.md or CLAUDE.md.

## How Cursor connects

Cursor connects through an MCP server entry in `~/.cursor/mcp.json` and the `codepanel` and `codepanel-setup` skills. The steps are in [Cursor](/docs/cursor).

## Cursor features for the review loop

| Feature | What the docs say | Source |
| --- | --- | --- |
| Maker | Anysphere | [cursor.com](https://cursor.com/terms-of-service) |
| Runs in | its own editor, a CLI, JetBrains IDEs through the Agent Client Protocol, cloud agents and an iOS app | [cursor.com](https://cursor.com/docs/get-started/quickstart), [page 2](https://cursor.com/docs/cli/overview), [page 3](https://cursor.com/docs/integrations/jetbrains), [page 4](https://cursor.com/docs/cloud-agent), [page 5](https://cursor.com/docs/cloud-agent/mobile) |
| MCP | stdio, SSE and Streamable HTTP servers, with OAuth, in `.cursor/mcp.json` or `~/.cursor/mcp.json` | [cursor.com](https://cursor.com/docs/mcp) |
| Hooks | Scripts that speak JSON over stdio, set in `hooks.json`; `stop` handles the agent finishing | [cursor.com](https://cursor.com/docs/hooks) |
| GitHub Actions | A documented workflow that installs the CLI and runs `agent -p` with `CURSOR_API_KEY` | [cursor.com](https://cursor.com/docs/cli/github-actions) |
| Headless mode | `agent -p "<prompt>"`, with `--force` to apply edits | [cursor.com](https://cursor.com/docs/cli/headless) |
| Rules files | [`.cursor/rules/*.mdc`](/agents/files/cursor-rules), [`AGENTS.md`](/agents/files/agents-md), [`CLAUDE.md`](/agents/files/claude-md) | [cursor.com](https://cursor.com/docs/rules), [page 2](https://cursor.com/help/customization/rules) |
| Models | Cursor's own models and models from OpenAI, Anthropic, Google and others, or your own key for OpenAI, Anthropic, Google, Azure OpenAI and AWS Bedrock. Prices: [OpenAI](/models/openai), [Anthropic](/models/anthropic), [Google](/models/google), [xAI](/models/xai), [Moonshot](/models/moonshot), [Z.ai](/models/zai). | [cursor.com](https://cursor.com/docs/models-and-pricing), [page 2](https://cursor.com/help/models-and-usage/api-keys) |

Each row was checked on 2026-10-17 against the maker's docs it cites. A feature with no row is one those docs don't describe.