Lets AI coding agents record, browse, and analyze real Chrome sessions with structured context including clicks, console logs, and network requests.
Also known as dundas/thinkrun, @thinkrun/mcp
- Stars
- 2
- Tools
- —
- Downloads, 30 days
- —
- Latest version
- 0.3.14
- Last commit
- Nov 9, 2026
- License
- MIT
What is ThinkRun MCP?
ThinkRun lets AI coding agents record, browse, and analyze real Chrome sessions with structured context including clicks, console logs, and network requests.
Summary generated from the project's README
About ThinkRun MCP server
ThinkRun provides AI coding agents with structured sight by recording real Chrome sessions. Instead of relying on static screenshots, developers can provide agents with time-synced context including clicks, console logs, network requests, and voice narration. This server works with popular MCP clients like Claude Code, Cursor, Cline, and Windsurf. It uses a native host binary to interact with your actual Chrome browser, meaning it can access your existing cookies and authenticated sessions without profile conflicts.
The server includes seven portable agent skills designed for specific development workflows. These include web-browse for programmatic navigation, ux-audit for generating structured UI reports, and prove-it for automated software-factory testing. Users can install the server via npx or install the CLI globally using npm. This setup allows an agent to navigate to local environments like localhost:3000 or analyze complex bugs through detailed Markdown or JSON reports rather than simple images.
What you can do with it
- 1Debug complex UI errorsProvide an agent with a recording containing console errors and network requests to fix a bug instead of just a screenshot.
- 2Automate software testingUse the prove-it skill to run test targets and receive a pass/fail table with supporting technical captures.
- 3Perform UX auditsInstruct an agent to walk through a product interface and generate a prioritized list of fixes.
- 4Verify visual changesCompare the appearance of a page before and after a code change using side-by-side captures.
Written from the project's README. Every use case is checked against a quote from it.
How to install ThinkRun MCP
Formats verified 10 Nov 2026Add to CursorAdd to GooseInstall ThinkRun MCP in Claude Code
$ claude mcp add thinkrun -- npx -y @thinkrun/mcpInstall ThinkRun MCP in Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"thinkrun": {
"command": "npx",
"args": [
"-y",
"@thinkrun/mcp"
]
}
}
}Install ThinkRun MCP in VS Code
.vscode/mcp.json
{
"servers": {
"thinkrun": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@thinkrun/mcp"
]
}
}
}Install ThinkRun MCP in VS Code CLI
$ code --add-mcp '{"name":"thinkrun","type":"stdio","command":"npx","args":["-y","@thinkrun/mcp"]}'Install ThinkRun MCP in Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"thinkrun": {
"command": "npx",
"args": [
"-y",
"@thinkrun/mcp"
]
}
}
}Install ThinkRun MCP in Codex
$ codex mcp add thinkrun -- npx -y @thinkrun/mcpInstall ThinkRun MCP in Codex config.toml
~/.codex/config.toml
[mcp_servers.thinkrun] command = "npx" args = ["-y", "@thinkrun/mcp"]
Install ThinkRun MCP in Gemini CLI
$ gemini mcp add -s user thinkrun npx -- -y @thinkrun/mcpInstall ThinkRun MCP in Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"thinkrun": {
"command": "npx",
"args": [
"-y",
"@thinkrun/mcp"
]
}
}
}Automated checks
Ran Nov 14, 2026Full report →Passed our automated checks. That is not a security audit: read the code and give the server the narrowest access it needs. How checks work
- Source repository linked and reachabledundas/thinkrun is public and reachablegithub.com/dundas/thinkrun ↗Pass
- Package names this repository as its source@thinkrun/mcp names dundas/thinkrun as sourcenpmjs.com/package/@thinkrun/mcp ↗Pass
- Build provenance (Sigstore attestation)No build attestation published for this releaseNot checked
- Open source license (OSI)MITspdx.org/licenses/MIT.html ↗Pass
- Commit or release in the last yearLast commit or release 2026-11-09github.com/dundas/thinkrun ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNo popular package one edit away from @thinkrun/mcpPass
- No critical advisory in the package or its dependencies112 resolved dependencies checked; no critical advisoryosv.dev ↗Pass
- No hidden instructions to the model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeNot checked
README
View on GitHubThinkRun — Screen Recorder for AI Coding Agents
Stop screenshotting bugs and typing paragraphs to explain them. ThinkRun records any browser session — or lets your agent record what it built — and turns it into structured context your AI coding agent can act on: every click, the console, network requests, your voice narration, and screenshots, all time-synced. Not just a screenshot. Not just a video.
Works with Claude Code, Cursor, Cline, Windsurf, and any MCP client.
Install
npx @thinkrun/mcp
Or install the CLI globally:
npm install -g @thinkrun/cli
The CLI automatically installs the native host binary, which lets AI tools control Chrome.
Add to your MCP config
{
"mcpServers": {
"thinkrun": {
"command": "npx",
"args": ["@thinkrun/mcp"]
}
}
}
Config location:
- Claude Code:
~/.claude/settings.json - Cursor:
.cursor/mcp.json - Cline: VS Code settings → Cline MCP Servers
Then ask your agent: "navigate to localhost:3000 and screenshot the checkout page" — or record the bug yourself and hand the agent the result.
Why ThinkRun
Other tools give your agent eyes. ThinkRun gives it structured sight — context it can parse in one pass instead of you re-explaining the page.
| ThinkRun | Playwright MCP | |
|---|---|---|
| Records a session into structured, replayable context | ✅ | ❌ drive-only |
| Turns a recording into an AI-ready report (clicks, console, network) | ✅ | ❌ |
| Shareable link, also as LLM-ready Markdown / JSON | ✅ | ❌ |
| Uses your real Chrome — your cookies & sessions | ✅ | ❌ headless only |
| Runs alongside your browser, no profile conflict | ✅ | ❌ fights for Chrome profile |
Agent skills for Claude Code, Cursor, Codex & Gemini CLI
Seven portable agent skills ship in this repo, mirrored for every major harness (.claude/skills/, .cursor/skills/, .codex/skills/, .gemini/skills/):
- prove-it — the prove step of a software-factory workflow: named test targets in, a pass/fail table with a screenshot, console and network capture per target out, posted to the PR; any non-pass row sends the change back to build.
- before-after — the ship step: two captures at the same viewport, one
| Before | After |table for the PR, no public paste host. - web-browse — browse the web programmatically: drive a real or cloud browser to navigate, interact, extract, and screenshot.
- ux-audit — walk through a product UI as a real user, screenshot every step, and produce a structured UX report with a prioritized fix list.
- thinkbrowse-cli — drive the browser from shell scripts via the
thinkrunCLI. - thinkbrowse-mcp — drive the browser through the ThinkRun MCP server's tools.
- browser-tool-benchmark — run the open Playwright CLI vs MCP benchmark on your own agent and model, and get tokens, cost, speed and pass rate per setup next to the published results.
Install without cloning — the repo is served as a live MCP server that auto-updates on every commit:
{ "mcpServers": { "thinkrun-skills": { "url": "https://mcp.skillsovermcp.com/mcp/dundas/thinkrun" } } }
Trigger accuracy is measured, not claimed. Every skill ships a 20-query trigger-eval fixture, and evals/BENCHMARK.md publishes the results: with the seven-skill catalog (v5, 2026-11-09), 140/140 on GPT-6 Luna, Claude Sonnet 5.5 and Claude Opus 5.5; with the six-skill catalog (v4, 2026-09-18), 120/120 on GPT-5.6-terra and Claude Fable 5 and 113/120 on Claude Sonnet 5 (its documented strict-MCP reading, not a crash); the earlier four-skill matrix scored 79–80/80 on 12 of 16 model configurations. Catalog-ablation runs show the browsing skills absorb each other's traffic when installed as a subset, and that prove-it fills a gap rather than duplicating a sibling (before-after folds into prove-it when absent). Reproduce with bun evals/run-eval.ts build.
Manual binary install
If you need to install the native host without the CLI, download the binary for your platform from the latest release and run:
chmod +x thinkbrowse-host-* && ./thinkbrowse-host-* --install
The native host binary keeps the
thinkbrowse-hostfilename for now — it's the same signed binary, unchanged by the rename.
FAQ
How do I give Claude Code (or Cursor) access to my browser?
Add ThinkRun as an MCP server — npx @thinkrun/mcp — and your agent can drive and read a real Chrome session, your cookies and logins included. See "Add to your MCP config" above.
How do I show Cursor a UI bug instead of pasting a screenshot?
Record the bug with ThinkRun. Instead of a flat image, your agent gets a structured artifact — the steps you took, the console errors, the failing network requests, and the DOM state at the moment it broke — so it fixes the actual problem instead of guessing from a picture.
Is there an MCP server that uses my real Chrome cookies and sessions?
Yes. ThinkRun runs against your real Chrome via a native host, so authenticated pages just work — no headless re-login, no separate profile, no fighting your browser for the Chrome profile.
How do I record a browser session for an AI coding agent?
Run npx @thinkrun/mcp (or npm i -g @thinkrun/cli) and record any session — or let the agent record what it built. ThinkRun turns it into an AI-ready report (clicks, console, network, narration, screenshots, all time-synced), available as a shareable link and as LLM-ready Markdown/JSON.
How do I give my AI coding agent context about what went wrong?
Hand it a ThinkRun recording. The structured "what happened" — actions, first bad signal, expected vs actual, and the element state — is exactly the context an agent needs to land a fix on the first try.
Benchmarks
- Playwright CLI vs MCP: 285 tests on Claude Sonnet 5.5, GPT-6 Luna and GPT-6.1 Sol, with the test shop, prompts, runners, results and the scripts that recompute every published number (video, blog post, ThinkRun analysis of the video)
More
ThinkRun MCP questions
How do I add ThinkRun MCP to Claude Code?
Run claude mcp add thinkrun -- npx -y @thinkrun/mcp in your terminal. Add --scope user to make it available in every project.
How do I add ThinkRun MCP to Cursor?
Open ~/.cursor/mcp.json and add the server under mcpServers, using the Cursor config shown in the install section above.
How do I give Claude Code (or Cursor) access to my browser?
Add ThinkRun as an MCP server using npx @thinkrun/mcp to allow your agent to drive and read a real Chrome session, including your cookies and logins.
How do I show Cursor a UI bug instead of pasting a screenshot?
Record the bug with ThinkRun to provide the agent with a structured artifact containing steps taken, console errors, failing network requests, and DOM state.
Is there an MCP server that uses my real Chrome cookies and sessions?
Yes, ThinkRun runs against your real Chrome via a native host, allowing authenticated pages to work without headless re-login or profile conflicts.
How do I record a browser session for an AI coding agent?
Run npx @thinkrun/mcp or install the CLI globally, then record a session to generate an AI-ready report with time-synced clicks, console, and network data.