$ loading_
Proxy Tavily search and extract APIs with key rotation and bearer auth.
This MCP tool is described as a proxy for Tavily search/extract APIs with multi-key rotation and bearer token authentication. While it is open source, the available materials are sparse, with no README and no visible adoption, so its actual network forwarding and credential handling deserve scrutiny.
The system metadata says no keys are required, but the description explicitly mentions support for multiple API keys and bearer token authentication, indicating the tool is designed to handle sensitive credentials. The materials do not explain credential source, storage, log redaction, or rotation boundaries, creating normal misconfiguration or leakage concerns.
As a proxy for Tavily search/extract, this tool would by design forward queries or extraction inputs to an external service. No specific host is listed, but the intended destination appears tied to Tavily APIs, which is expected network egress. Due to missing documentation, it is unclear whether it also sends data to other endpoints or logs request contents.
The system flags code execution, meaning it runs as a local MCP service process. The available materials do not show requests for unusual system privileges, nor any execution capabilities beyond its stated proxy role. With limited information, this should be treated as the normal caution level for a local service process.
As a proxy service, it would at least handle user-submitted search terms, URLs, or extraction text in requests. The materials do not say whether it reads/writes local files, caches responses, stores logs, or exposes request history, so data exposure and retention are opaque, though no clear over-privileged access is shown.
A positive sign is that there is a public source repository that can be audited. However, it comes from a third-party registry, has no declared license, no README, zero stars, and unknown maintenance status, so overall trust and maturity are limited. No explicit malicious red flags are visible, but supply-chain transparency is weak.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "tavily-mcp-proxy" yet — see the docs or source repo.
Use Tavily to search for materials about “AI search agents” from the last 7 days, then list the 5 most relevant results with source links and a one-sentence takeaway.
Return 5 relevant search results with links, summaries, and takeaways.
Extract the main content from this web page and summarize it into 3 bullets: https://example.com/article
Provide a content summary and 3 key points from the page.
For these three topics, search authoritative sources and verify the facts: A, B, and C. Provide 2 sources and a short conclusion for each topic.
Output verification results, sources, and short conclusions by topic.
Use Tavily via MCP for search, extraction, crawling, and web research workflows.
Search, extract, map, and crawl live web data through the Tavily API.
Access real-time web search, extraction, mapping, and crawling with optional PII protection.
Search the web, extract content, crawl sites, and support AI research.
Search and scrape web content with automatic Tavily-to-Firecrawl fallback.
Enable web search in AI workflows via the Tavily-powered MCP server.