mcp-nexus
by Greenchannel
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| HOST | No | The host address for the server to listen on. Default: 0.0.0.0 | |
| PORT | No | The port for the server to listen on. Default: 8787 | |
| DATABASE_URL | No | Connection string for the database. Default: file:./tavily_bridge.db | |
| BRAVE_API_KEY | No | A Brave Search API key. Default: '' | |
| BRAVE_MAX_QPS | No | Max requests per second to Brave API. Default: 1 | |
| BRAVE_OVERFLOW | No | Behavior when the request queue is full: fallback_to_tavily, queue, or error. Default: fallback_to_tavily | |
| ADMIN_API_TOKEN | Yes | Bearer token for accessing the Admin API. | |
| MCP_COOLDOWN_MS | No | Cooldown period in milliseconds for an upstream API key after a failure. Default: 60000 | |
| MCP_MAX_RETRIES | No | Maximum number of retries for failed upstream requests. Default: 2 | |
| ENABLE_QUERY_AUTH | No | If true, also allow passing the MCP client token via query string. Default: false | |
| BRAVE_MAX_QUEUE_MS | No | Max time a request can wait in the queue before failing or falling back to Tavily. Default: 30000 | |
| SEARCH_SOURCE_MODE | No | Defines search behavior: tavily_only, brave_only, combined, or brave_prefer_tavily_fallback. Default: brave_prefer_tavily_fallback | |
| BRAVE_HTTP_TIMEOUT_MS | No | Per-request HTTP timeout for the Brave API. Default: 20000 | |
| BRAVE_MIN_INTERVAL_MS | No | Overrides BRAVE_MAX_QPS with a fixed minimum interval. Default: '' | |
| KEY_ENCRYPTION_SECRET | Yes | A 32-byte (256-bit) secret key used for encrypting and decrypting upstream API keys stored in the database. | |
| TAVILY_USAGE_LOG_MODE | No | Log level for Tavily tool usage: none, hash, preview, or full. Default: preview | |
| TAVILY_USAGE_HASH_SECRET | No | Optional secret for creating a keyed HMAC-SHA256 hash. Default: '' | |
| TAVILY_USAGE_SAMPLE_RATE | No | Optional sampling rate. Default: '' | |
| MCP_RATE_LIMIT_PER_MINUTE | No | Max requests per minute per client token. Default: 60 | |
| TAVILY_CREDITS_COOLDOWN_MS | No | Cooldown duration. Default: 300000 | |
| TAVILY_CREDITS_CACHE_TTL_MS | No | Cache TTL. Default: 60000 | |
| TAVILY_USAGE_RETENTION_DAYS | No | Optional retention period. Default: '' | |
| TAVILY_CREDITS_MIN_REMAINING | No | Credit threshold for cooldown. Default: 1 | |
| TAVILY_KEY_SELECTION_STRATEGY | No | Strategy for picking an upstream Tavily key: round_robin or random. Default: round_robin | |
| TAVILY_CREDITS_REFRESH_LOCK_MS | No | Lock duration. Default: 15000 | |
| MCP_GLOBAL_RATE_LIMIT_PER_MINUTE | No | Max requests per minute across all clients. Default: 600 | |
| TAVILY_USAGE_CLEANUP_PROBABILITY | No | Probability of cleanup trigger. Default: 0.001 | |
| TAVILY_CREDITS_REFRESH_TIMEOUT_MS | No | Timeout. Default: 5000 | |
| ADMIN_KEY_REVEAL_RATE_LIMIT_PER_MINUTE | No | Max key reveal attempts per minute in Admin UI. Default: 20 |
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
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
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/Greenchannel/mcp-nexus'
If you have feedback or need assistance with the MCP directory API, please join our Discord server