Odysseus Web MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| SERPER_API_KEY | No | API key for Serper search | |
| TAVILY_API_KEY | No | API key for Tavily search | |
| WEB_MCP_DATA_DIR | No | Directory for persistent state; defaults to ~/.local/share/odysseus-web-mcp | |
| DATA_BRAVE_API_KEY | No | API key for Brave search | |
| WEB_MCP_RENDER_ENABLED | No | Set to 'true' to enable JavaScript rendering via Playwright | |
| WEB_MCP_SEARCH_PROVIDER_CHAIN | No | Comma-separated list of search providers to try in order |
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": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| web_searchA | Discover public web sources for focused questions. Returns provider outcomes, ranked source URLs, snippets, evidence_id, and truthful empty/partial status. Use web_fetch to read a known URL. |
| web_fetchA | Read one known public URL or a bounded batch of URLs. Enforces public-URL SSRF checks, redirect and byte limits, returns extracted text, quality signals, discovered links, evidence_id, and cursor continuation when content is long. |
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 2 tools
web_search and web_fetch are clearly distinct: one discovers pages via queries, the other retrieves a known URL. No functional overlap.
Both tools follow a consistent web_verb pattern, making the surface predictable and easy to navigate.
Only two tools are exposed, which is slightly below the typical 3-15 range, but each tool is essential and the set is appropriately narrow for a web search/fetch server.
Search and fetch cover the core web access workflow, including pagination/cursor handling, but the surface is minimal and may lack advanced retrieval features such as site-specific extraction or structured data parsing.