research-mcp
by vvzvlad
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| RETRIES | No | Number of retries per provider call. | 2 |
| LOG_FILE | No | Path to the persistent log file. | data/research-mcp.log |
| MCP_HOST | No | Host to bind the MCP server. | 0.0.0.0 |
| MCP_PORT | No | Port to bind the MCP server. | 8000 |
| LOG_LEVEL | No | Logging level (e.g., DEBUG, INFO, WARNING). | INFO |
| EXA_API_KEY | No | API key for the Exa search service. | |
| SEARXNG_URL | No | Base URL of the SearXNG instance (e.g., http://localhost:8080). | |
| CRAWL4AI_URL | No | Base URL of the Crawl4AI instance. | |
| JINA_API_KEY | No | Optional API key for Jina AI reader. | |
| LOG_ROTATION | No | Maximum size for log file rotation. | 20 MB |
| LOG_RETENTION | No | Retention period for rotated log files. | 14 days |
| CRAWL4AI_TOKEN | No | Token for Crawl4AI authentication. | |
| SERPER_API_KEY | No | API key for the Serper.dev search service. | |
| REQUEST_TIMEOUT | No | Timeout (seconds) for upstream requests. | 30 |
| TAVILY_1_API_KEY | No | API key for the first Tavily instance. | |
| TAVILY_2_API_KEY | No | API key for the second Tavily instance (fallover). | |
| FIRECRAWL_API_KEY | No | API key for the Firecrawl service. | |
| FALLBACK_MIN_CHARS | No | Minimum characters for a read result to be accepted. | 500 |
| READ_PAGES_CONCURRENCY | No | Concurrent page reads for read_pages tool. | 5 |
Capabilities
Server capabilities have not been inspected yet.
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
No tools | |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/vvzvlad/research-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server