CortexScout
by cortex-works
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PORT | No | Listening port for the HTTP server binary (cortex-scout). | 5000 |
| RUST_LOG | No | Log level. Keep warn for MCP stdio — info floods stderr and confuses MCP clients. | warn |
| MAX_LINKS | No | Max links followed per page crawl. | 100 |
| LANCEDB_URI | No | Directory path for persistent research memory. Omit to disable. | |
| IP_LIST_PATH | No | Optional path to ip.txt (one proxy per line: http://, socks5://). Leave unset to disable proxy support entirely, or point at an empty file to keep proxy tools available but inactive by default. | |
| OPENAI_API_KEY | No | API key for LLM synthesis. Omit for key-less local endpoints (Ollama). | |
| OUTBOUND_LIMIT | No | Max concurrent outbound HTTP connections. | 32 |
| SEARCH_ENGINES | No | Active engines (comma-separated). | google,bing,duckduckgo,brave |
| MODEL2VEC_MODEL | No | HuggingFace model ID or local path for embedding (e.g. minishlab/potion-base-8M). | |
| OPENAI_BASE_URL | No | OpenAI-compatible endpoint (OpenRouter, Ollama, LM Studio, etc.). | https://api.openai.com/v1 |
| CHROME_EXECUTABLE | No | Override path to Chromium/Chrome/Brave binary. | |
| CORTEX_SCOUT_PORT | No | Listening port for the HTTP server binary (cortex-scout). | 5000 |
| HTTP_TIMEOUT_SECS | No | Per-request read timeout (seconds). | 30 |
| MAX_CONTENT_CHARS | No | Max characters returned per scraped page. | 10000 |
| PROXY_SOURCE_PATH | No | Optional path to proxy_source.json (used by proxy_control grab). | |
| SEARCH_CDP_FALLBACK | No | Retry search engine fetches via native Chromium CDP when blocked. | true |
| DEEP_RESEARCH_ENABLED | No | Set 0 to disable the deep_research tool at runtime. | 1 |
| SEARCH_TIER2_NON_ROBOT | No | Set 1 to allow hitl_web_fetch as last-resort search escalation. | |
| DEEP_RESEARCH_LLM_MODEL | No | Model identifier (must be supported by the endpoint). | gpt-4o-mini |
| DEEP_RESEARCH_SYNTHESIS | No | Set 0 to skip LLM synthesis (search+scrape only). | 1 |
| HTTP_CONNECT_TIMEOUT_SECS | No | TCP connect timeout (seconds). | 10 |
| CORTEX_SCOUT_MEMORY_DISABLED | No | Set 1 to disable memory even when LANCEDB_URI is set. | 0 |
| SEARCH_MAX_RESULTS_PER_ENGINE | No | Results per engine before merge/dedup. | 10 |
| DEEP_RESEARCH_SYNTHESIS_MAX_TOKENS | No | Max tokens for synthesis response. Use 4096+ for large-context models. | 1024 |
| DEEP_RESEARCH_SYNTHESIS_MAX_SOURCES | No | Max source documents fed to LLM synthesis. | 8 |
| DEEP_RESEARCH_SYNTHESIS_MAX_CHARS_PER_SOURCE | No | Max characters extracted per source for synthesis. | 2500 |
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
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
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- 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/cortex-works/cortex-scout'
If you have feedback or need assistance with the MCP directory API, please join our Discord server