researcher
by mareurs
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| RUST_LOG | No | Log level filter | info |
| BIND_ADDR | No | HTTP server bind address | 0.0.0.0:3000 |
| FB_COOKIE | No | Cookie header for facebook.com | |
| LLM_MODEL | No | Model name sent in requests | Qwen3.5-4B-Q4_K_M |
| EXA_API_KEY | No | Exa API key — used for academic profile (empty = disabled) | |
| LLM_API_KEY | No | Set to sk-... for OpenAI | no-key-needed |
| SEARXNG_URL | No | SearXNG instance URL | http://localhost:4000 |
| LLM_BASE_URL | No | Any OpenAI-compatible endpoint | http://localhost:8080/v1 |
| ADZUNA_APP_ID | No | Adzuna API app ID (job search) — free at developer.adzuna.com | |
| BRAVE_API_KEY | No | Brave Search API key (empty = disabled; default fallback for all profiles) | |
| ADZUNA_APP_KEY | No | Adzuna API key | |
| ADZUNA_COUNTRY | No | Adzuna country code (us, gb, de, fr, …) | us |
| EMBED_BASE_URL | No | TEI embed endpoint (e.g. http://localhost:8081) | |
| LLM_FAST_MODEL | No | Model name for fast LLM | Qwen3.5-4B-Q4_K_M |
| LLM_MAX_TOKENS | No | Max tokens per LLM call | 4096 |
| MAX_PAGE_CHARS | No | Max characters extracted per page | 8000 |
| TAVILY_API_KEY | No | Tavily API key — used for news profile (empty = disabled) | |
| TWITTER_COOKIE | No | Cookie header for twitter.com / x.com | |
| DEDUP_THRESHOLD | No | Cosine similarity cutoff for deduplication | 0.92 |
| LINKEDIN_COOKIE | No | Cookie header for linkedin.com | |
| LLM_FAST_STAGES | No | Pipeline stages routed to fast LLM | planner,summarizer,publisher |
| LLM_TEMPERATURE | No | Generation temperature | 0.3 |
| RERANK_BASE_URL | No | TEI rerank endpoint (e.g. http://localhost:8082) | |
| INSTAGRAM_COOKIE | No | Cookie header for instagram.com | |
| LLM_FAST_API_KEY | No | Fast LLM API key; empty = inherit LLM_API_KEY | |
| MIN_TEXT_DENSITY | No | Minimum text/HTML density ratio | 0.05 |
| LLM_FAST_BASE_URL | No | Fast LLM endpoint; empty = use heavy backend | |
| MIN_CONTENT_WORDS | No | Minimum word count per page | 100 |
| MAX_SEARCH_QUERIES | No | Sub-questions the planner generates | 4 |
| LLM_FAST_MAX_TOKENS | No | Max tokens for fast model | 4096 |
| MAX_SOURCES_PER_QUERY | No | Pages scraped per query | 4 |
| RERANK_QUALITY_WEIGHT | No | Content quality weight | 0.1 |
| STRIP_THINKING_TOKENS | No | Strip <think>...</think> from Qwen3 responses | true |
| RERANK_AUTHORITY_WEIGHT | No | Domain authority weight | 0.2 |
| RERANK_RELEVANCE_WEIGHT | No | Cross-encoder score weight | 0.7 |
| SEARCH_RESULTS_PER_QUERY | No | Results fetched per sub-question | 8 |
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/mareurs/researcher'
If you have feedback or need assistance with the MCP directory API, please join our Discord server