Skip to main content
Glama
Biogod2020
by Biogod2020

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
DSH_WEB_PROXYNoHTTP/HTTPS/SOCKS proxy. The process uses trust_env=False and does not inherit HTTP_PROXY
DSH_BING_SEARCH_URLNoOverride Bing HTML endpoint only when set to a non-default value (tests). Otherwise the host is chosen by languagehttps://www.bing.com/search
DSH_WEB_CONCURRENCYNoMaximum in-process concurrent requests8
DSH_WEB_IMPERSONATENocurl_cffi browser fingerprintchrome
DSH_WEB_MAX_REDIRECTSNoMaximum redirects8
DSH_WEB_MAX_BODY_BYTESNoMaximum body size for open5242880
DSH_BING_MAX_BODY_BYTESNoMaximum search-page body size2097152
DSH_WEB_TIMEOUT_SECONDSNoTransfer timeout20
DSH_WEB_CACHE_TTL_SECONDSNoPage cache TTL600
DSH_BING_CACHE_TTL_SECONDSNoSearch cache TTL90
DSH_WEB_CONNECT_TIMEOUT_SECONDSNoConnect timeout8

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

CapabilityDetails
tools
{
  "listChanged": false
}
prompts
{
  "listChanged": false
}
resources
{
  "subscribe": false,
  "listChanged": false
}
experimental
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
searchA

Search the public web. DuckDuckGo is tried first when reachable; Bing is the fallback.

Chinese queries / zh-* markets use cn.bing.com. Read quality_label: poor means the titles are unrelated or first-token junk — do not treat them as answers.

Args: query: Compact concrete nouns plus the qualifier that uniquely identifies the subject. "复旦光华楼" is better than "光华楼" — the extra place/institution is necessary, not padding. Do not write whole sentences. If a compact query is still ambiguous or hits the wrong entity, write more (place, institution, year, type). For a person plus a paper, search the author name first. count: Number of organic results to return, from 1 to 20. offset: Result offset for pagination, from 0 to 100. market: Locale such as en-US or zh-CN. Chinese text should use zh-CN. safe_search: SafeSearch level (used when Bing is the engine).

search_imagesA

Search image indexes and rank results with pure text so vision is not required.

auto (default) tries Bing Images first and falls back to Wikimedia Commons when the top text score is below ~40, so one call yields a ranked set. bing_images parses Bing Images metadata (original URL / thumbnail / source page / title). commons queries Wikimedia Commons, a curated and licence-clear platform. Every result carries a 0-100 text score, a domain hint and explainable signals; pick the highest score, treat scores below ~40 as unverified, and optionally verify with find/open on the source page before downloading.

Args: query: What the image should depict. Compact concrete nouns plus the qualifier that uniquely identifies the subject (e.g. "复旦光华楼", "台州城墙"). "复旦光华楼" is better than "光华楼". Do not write whole sentences. If a compact query is still ambiguous or hits the wrong entity, write more (place, institution, year, type). count: Number of ranked image results to return, from 1 to 20. market: Locale such as en-US or zh-CN (Bing Images; Commons is language-neutral). provider: auto (default), bing_images, or commons.

openA

Fetch a public HTTP(S) page with curl_cffi and return cleaned readable text.

Use after search when result snippets are insufficient. Private/local addresses are rejected, redirect targets use curl_cffi safe-follow mode, and response bytes are capped.

findA

Find a literal phrase in a page and return compact context windows around matches.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

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/Biogod2020/dsh-bing-search'

If you have feedback or need assistance with the MCP directory API, please join our Discord server