tavily-firecrawl-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| TAVILY_API_KEY | Yes | Tavily API key for web search and extraction | |
| WEB_FETCH_PROXY | No | Optional proxy URL for outbound requests | |
| FIRECRAWL_API_KEY | No | Optional Firecrawl API key for fallback web fetching |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
Web fetching (Tavily primary, Firecrawl automatic fallback) and web search (Tavily). Use web_fetch to retrieve a page's content; use web_search to find pages.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| web_fetchA | Fetch and extract the content of a web page as markdown (or plain text). Tries Tavily first; on failure or empty result, automatically falls back to Firecrawl. Requires TAVILY_API_KEY and/or FIRECRAWL_API_KEY in the server's environment. |
| web_searchB | Search the web and return titles, URLs, and snippets. Tries Tavily first (optionally with a synthesized answer); on failure falls back to Firecrawl Search. Requires TAVILY_API_KEY and/or FIRECRAWL_API_KEY in the server's environment. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 2 tools
web_search (discovery of titles/URLs/snippets) and web_fetch (retrieval of full page content) are clearly distinct actions on different outputs. There is no plausible scenario where an agent would confuse which one to call first.
Both tools follow an identical 'web_<verb>' snake_case pattern, making the naming perfectly predictable. Adding a third tool would be trivially guessable (e.g., web_crawl).
Two tools is slightly below the typical 3-15 sweet spot, but for a focused web-access server each tool earns its place and there is no filler. The scope is intentionally minimal rather than padded.
Search plus fetch covers the core read-only lifecycle of web access, including provider fallback for reliability. Minor gaps exist (no crawl/site-mapping, no batch or structured extract), but agents can work around these by chaining search into fetch.