web-search-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| JINA_API_KEY | No | API key for Jina Reader. Optional; increases fetch_page rate limits. | |
| FREE_BACKENDS | No | Comma-separated list of backend engines for the free tier. Invalid names are filtered with a warning. Defaults to 'auto'. | auto |
| SERPER_API_KEY | No | API key for Serper.dev (Google SERP). Optional; enables the Serper provider. | |
| TAVILY_API_KEY | No | API key for Tavily. Optional; enables the Tavily provider. | |
| FETCH_CACHE_TTL | No | TTL for fetch cache in seconds. | 3600 |
| FREE_SEARCH_RETRIES | No | Number of retries for the free search tier. | 2 |
| WEB_SEARCH_CACHE_DB | No | Path to the cache.db file. Defaults to the OS cache directory for web-search-mcp. | OS cache directory (platformdirs.user_cache_dir("web-search-mcp")) |
| WEB_SEARCH_CACHE_TTL | No | TTL for search cache in seconds. | 3600 |
| MAX_RESULTS_PER_DOMAIN | No | Maximum number of URLs kept per domain in search results. | 2 |
| WEB_SEARCH_STATS_MAX_ROWS | No | Maximum number of metrics records stored in SQLite. | 1000 |
| WEB_SEARCH_CACHE_MAX_ENTRIES | No | Maximum number of cache entries to keep before trimming. | 500 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| logging | {} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| extensions | {
"io.modelcontextprotocol/ui": {}
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| web_searchB | Tìm kiếm trên web và trả về danh sách kết quả. |
| fetch_pageA | Lấy nội dung đầy đủ của một trang web dưới dạng markdown. |
| search_statsA | Xem provider nào hay lỗi và chậm bao nhiêu, trong N giờ gần nhất. Dùng khi search trả kết quả kém hoặc chậm, để biết nên đặt API key hay không. |
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 3 tools
Each tool has a clearly distinct purpose: web_search returns result lists, fetch_page retrieves full page content, and search_stats monitors provider performance. There is no meaningful overlap between them.
All tool names follow a consistent verb_noun snake_case pattern: web_search, fetch_page, and search_stats. The naming is predictable and easy for an agent to parse.
Three tools is well-scoped for a web search server: search, fetch, and operational diagnostics. Each tool earns its place with no redundancy.
The core workflow of searching the web and then fetching full page content is fully covered, and search_stats adds useful operational insight. There are no obvious dead ends or missing critical operations for this domain.