$ loading_
Search and fetch web pages as plain text with lower token usage.
The materials indicate this MCP tool mainly performs web search and web fetch via a local tbro terminal browser and returns plain text; no secret requirement or clearly excessive data permission is stated. Overall risk appears relatively low, but local execution and outbound network access plus sparse documentation, missing license, and unknown maintenance warrant cautious use.
The materials explicitly state there are no required secrets or environment variables. No API token, account credential, or other highly sensitive authentication material is requested, so credential exposure appears limited.
The described web_search and web_fetch features indicate the tool will access external websites or search services on the user's behalf and retrieve page text. Although no fixed remote endpoint is declared, user queries and target URLs may be sent to relevant sites, which is normal outbound behavior for this type of tool.
The system flags this tool as executes-code, and the description says it relies on the tbro terminal browser, implying it starts or invokes a local program/process to perform fetching and searching. This is a common execution surface for MCP tools, and the provided materials do not show any additional dangerous system privilege request.
The available materials only state that the tool returns plain text from web content. They do not state a need to read local files, write to disk, access databases, or request data permissions unrelated to web retrieval; based on the known facts, no obvious expansion of data access is shown.
The project has a public GitHub repository and is open source, which is a meaningful risk-reducing factor. However, it comes from a third-party registry, has 0 stars, no declared license, unknown maintenance status, and no README, so supply-chain transparency and maturity are only moderate; code and dependency review is advisable before use.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "tbro-mcp" yet — see the docs or source repo.
Use tbro-mcp to fetch this page and return plain text only. Keep the title and main paragraphs, and remove navigation, ads, and irrelevant links: https://example.com/article
Returns the cleaned main body text of the page, ready for summarization, analysis, or LLM input.
Use tbro-mcp to search for "Python async queue best practices" and list the top 5 most relevant results with title, URL, and a one-sentence summary.
Returns a structured list of search results for quick review and source selection.
First search for "EU AI Act official guidance", then fetch the full text of the 3 most relevant results, extract the key points from each, and include source links.
Produces a research brief combining key points and links from multiple fetched sources.
Let AI browse via your real Chrome for extraction and multi-step workflows.
Control a browser with AI for automation, extraction, interception, and screenshots.
Search Google and view web pages with bot-resistant automation.
Search the web with natural language and quickly gather external information.
Fetch web pages as markdown for LLM reading, analysis, and automation.
Browse websites in terminal and extract content and links for analysis.