Skip to main content
Glama
DeepSeek Search Everywhere MCP

DeepSeek Search Everywhere MCP

deepseek_search_many

Read-onlyIdempotent

Run up to five web searches in parallel to cover multiple angles at once and get ranked results per query in one call.

Instructions

Run up to 5 web searches in parallel and get ranked results per query in one call.

Use it to cover several angles, phrasings or sub-topics at once. Each query <= 128 chars; $0.00075 per query; page text across all queries defaults to 20,000 chars. Keeps input order, dedupes repeated queries, runs the first 5 unique ones and lists the rest in skipped_queries. Untrusted content is sanitized.

When NOT to use: one query (deepseek_search), known URLs (deepseek_fetch_many), fields as JSON (deepseek_extract_many), or a synthesized answer (deepseek_research).

Returns: {"ok": True, "results": [{"query": str, "results": [...]}], "partial": bool, "cost_usd": float}.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "queries": "<queries>"
    },
    "name": "deepseek_search_many"
  }
}

Input Schema

NameRequiredDescriptionDefault
langNoRestrict results to a language code such as 'en' or 'pt-BR' (default None)
afterNoFilter results published on or after YYYY-MM-DD date (default None)
countNoNumber of ranked results to return (1-50, default 5)
beforeNoFilter results published on or before YYYY-MM-DD date (default None)
contentNoResult format: 'snippets' (default: snippet, else summary capped at 400 chars) or 'full' (gap-filled text)snippets
queriesYesSearch queries to run concurrently (1-5 queries run, extras skipped; each <= 128 chars)
refreshNoBypass local cache and fetch fresh results from API (default False)
rewriteNoTranslate the query to English with deepseek-flash; skipped when lang is set (default False)
deadlineNoOverall timeout deadline in seconds; completed results returned if reached
exact_phrasesNoPhrases that must appear verbatim in search results (default None)
exclude_termsNoTerms that must not appear in search results (default None)
exclude_domainsNoExclude specific domains from search results (default None)
include_domainsNoRestrict search to specific domains or sites (default None)
max_total_charsNoMax total characters of page text across all queries (default 20000)

Output Schema

NameRequiredDescriptionDefault

No arguments

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed5 schema fields changedv3.3.0
    • changedInput schema / properties / lang / description
      Previous value: -"Restrict results to ISO language code like 'en', 'de', 'pl' (default None)"New value: +"Restrict results to a language code such as 'en' or 'pt-BR' (default None)"
    • changedInput schema / properties / max_total_chars / default
      Previous value: -30000New value: +20000
    • changedInput schema / properties / max_total_chars / description
      Previous value: -"Max total characters of page text across results (default 30000)"New value: +"Max total characters of page text across all queries (default 20000)"
    • changedInput schema / properties / queries / description
      Previous value: -"Search queries to execute concurrently (1-5 queries, each <= 128 chars)"New value: +"Search queries to run concurrently (1-5 queries run, extras skipped; each <= 128 chars)"
    • changedInput schema / properties / rewrite / description
      Previous value: -"Rewrite search query via deepseek-flash for better retrieval (default False)"New value: +"Translate the query to English with deepseek-flash; skipped when lang is set (default False)"
  2. Changed20 schema fields changedv3.0.0
    • addedInput schema / properties / after / description
      Added value: +"Filter results published on or after YYYY-MM-DD date (default None)"
    • addedInput schema / properties / before / description
      Added value: +"Filter results published on or before YYYY-MM-DD date (default None)"
    • addedInput schema / properties / content
      Added value: +{
      +  "default": "snippets",
      +  "description": "Result format: 'snippets' (default: snippet, else summary capped at 400 chars) or 'full' (gap-filled text)",
      +  "enum": [
      +    "snippets",
      +    "full"
      +  ],
      +  "title": "Content",
      +  "type": "string"
      +}
    • addedInput schema / properties / count / description
      Added value: +"Number of ranked results to return (1-50, default 5)"
    • addedInput schema / properties / deadline / description
      Added value: +"Overall timeout deadline in seconds; completed results returned if reached"
    • removedInput schema / properties / dedupe
      Removed value: -{
      -  "default": true,
      -  "title": "Dedupe",
      -  "type": "boolean"
      -}
    • addedInput schema / properties / exact_phrases / description
      Added value: +"Phrases that must appear verbatim in search results (default None)"
    • addedInput schema / properties / exclude_domains / description
      Added value: +"Exclude specific domains from search results (default None)"
    • addedInput schema / properties / exclude_terms / description
      Added value: +"Terms that must not appear in search results (default None)"
    • addedInput schema / properties / include_domains / description
      Added value: +"Restrict search to specific domains or sites (default None)"
    • addedInput schema / properties / lang / description
      Added value: +"Restrict results to ISO language code like 'en', 'de', 'pl' (default None)"
    • removedInput schema / properties / localization
      Removed value: -{
      -  "anyOf": [
      -    {
      -      "type": "string"
      -    },
      -    {
      -      "type": "null"
      -    }
      -  ],
      -  "default": null,
      -  "title": "Localization"
      -}
    • removedInput schema / properties / max_chars_per_result
      Removed value: -{
      -  "default": 4000,
      -  "title": "Max Chars Per Result",
      -  "type": "integer"
      -}
    • addedInput schema / properties / max_total_chars / description
      Added value: +"Max total characters of page text across results (default 30000)"
    • removedInput schema / properties / need_content
      Removed value: -{
      -  "default": false,
      -  "title": "Need Content",
      -  "type": "boolean"
      -}
    • addedInput schema / properties / queries / description
      Added value: +"Search queries to execute concurrently (1-5 queries, each <= 128 chars)"
    • addedInput schema / properties / refresh / description
      Added value: +"Bypass local cache and fetch fresh results from API (default False)"
    • addedInput schema / properties / rewrite / description
      Added value: +"Rewrite search query via deepseek-flash for better retrieval (default False)"
    • removedInput schema / properties / strip_boilerplate
      Removed value: -{
      -  "default": true,
      -  "title": "Strip Boilerplate",
      -  "type": "boolean"
      -}
    • removedInput schema / properties / summary_chars
      Removed value: -{
      -  "anyOf": [
      -    {
      -      "type": "integer"
      -    },
      -    {
      -      "type": "null"
      -    }
      -  ],
      -  "default": null,
      -  "title": "Summary Chars"
      -}
  3. Addedv2.3.0

TDQS

A4.6/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Beyond the readOnly/openWorld/idempotent annotations, it discloses per-query cost ($0.00075), the 20,000-char page-text budget, the 128-char query cap, dedupe of repeated queries, the first-5-unique execution rule with skipped_queries reporting, input-order preservation, partial-result behavior, and that untrusted content is sanitized. These are exactly the traits an agent needs before paying for a batch call.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the action, then scoped by a usage paragraph and a clearly delimited 'When NOT to use' and 'Returns' block. Every sentence carries information, though the Returns block partly duplicates the existing output schema.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With 14 parameters, rich annotations, and an output schema, the description supplies the cost, timeout/partial semantics, limits, and result grouping an agent needs to call it correctly. Nothing material is left to inference.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so baseline is 3, but the description adds meaning the schema does not: the dedupe/ordering/skipped_queries semantics of the queries array and the default 20,000-char aggregate budget. It does not explain the interaction between count, content, and max_total_chars, which keeps it short of a 5.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The opening sentence gives a specific verb (run web searches in parallel), a hard scope limit (up to 5), and the return shape (ranked results per query, single call). It is immediately distinguishable from deepseek_search by the batch dimension.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It states a positive use case (cover several angles/phrasings/sub-topics at once) and an explicit 'When NOT to use' block naming deepseek_search, deepseek_fetch_many, deepseek_extract_many, and deepseek_research with the condition that selects each. The only gap is that the positive guidance is thinner than the negative routing.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.