$ loading_
Control browsers with natural language for research, scraping, testing, and workflows.
This is an open-source MIT-licensed browser automation MCP that requires no API keys and declares no fixed remote endpoints. However, it has local execution and browser-control capabilities, and its description mentions high-privilege/evasive features such as 'full DevTools access' and 'stealth mode', so it should be used with caution rather than being labeled high risk from the current evidence alone.
The material explicitly states that no keys or environment variables are required. No API tokens, account passwords, or other sensitive credentials are requested, so the credential exposure surface appears limited based on the provided facts.
Although no fixed remote endpoint is declared, this is a browser automation tool and will likely visit arbitrary websites on the user's behalf and exchange data with external sites during normal use. The material does not define data egress boundaries, domain restrictions, or logging behavior, so users should be cautious about browser content being sent to third-party websites.
The system flags it as executes-code, and the description claims 40 tools, natural-language browser control, and full DevTools access, indicating it can start/drive local browser processes and use powerful debugging features. This is a typical high-privilege capability for this kind of MCP, so it should be run in a controlled environment.
The material does not clearly specify file-system read/write scope, but browser automation with DevTools access can typically interact with page content, session state, downloads, and browser-held data. There is no explicit request for system privileges clearly unrelated to its stated function, so caution is more appropriate than high risk at this stage.
Positive signals include being open source, auditable, and MIT-licensed. However, it comes from a third-party registry, has 0 GitHub stars, unknown maintenance status, no README content, and includes marketing/evasive wording such as 'stealth mode' and 'ghost effects,' which weakens supply-chain trust. The current evidence is insufficient to label it high risk outright, but source and dependency review is recommended first.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "IRA-RESEARCHER" yet — see the docs or source repo.
Open 5 competitor websites, extract their pricing, key value propositions, and signup entry points, then organize them into a comparison table.
A structured comparison table with pricing, feature highlights, and key page entry points.
Visit my staging site, check the login flow, form submission, and console errors, then output a test results summary.
A test summary describing successful steps, failure points, and detected browser errors.
Open a list of target webpages, scrape the title, publish date, author, and content summary, deduplicate the results, and export them as JSON.
A deduplicated structured data file ready for analysis or database import.
Automate browsers, extract web data, and inspect network requests efficiently.
Automate browser navigation, interaction, and web data extraction for online tasks.
Automate complex web booking and browsing workflows for travel and ticketing tasks.
Automate web browsing, interaction, and data extraction with a real Chrome fingerprint.
Automate browser tasks, capture console logs, and take screenshots for web workflows.
Automate browser actions like navigation, form filling, clicking, screenshots, and tab management.