Skip to main content
Glama
DeepSeek Search Everywhere MCP

DeepSeek Search Everywhere MCP

deepseek_research

Read-onlyIdempotent

Research any topic across multiple web sources and return a compact cited report in one call, with numbered citations matched to sources instead of raw pages filling your context.

Instructions

Research a topic across the web and get a cited report in one call.

The server searches several angles, reads 6-24 sources and returns a compact report whose [n] citations match the ids in the sources array, instead of filling your context with raw pages. About $0.003 (depth "low") to $0.03 ("max"). depth: "low" quick overview, "medium" (default) most questions, "high" decisions with several viewpoints, "max" broad or contested topics. sources="academic" favours peer-reviewed papers, "technical" official docs, specs and repos. Numbers missing from their cited source are marked [unverified] (unverified="drop" removes those sentences). If status is "running", call deepseek_check_research with the job_id; do not start the same research again. With synthesis="client" you get the ranked sources in pages and write the report yourself; the server then verifies its citations and numbers. At most 3 jobs run at once.

When NOT to use: one fact or query (deepseek_search), several independent queries (deepseek_search_many), known URLs (deepseek_fetch), or fields as JSON (deepseek_extract).

Returns: {"ok": True, "status": "done"|"running"|"awaiting_report", "job_id"?, "report": str, "sources": [{id, title, url, date, doi, pmid, pmcid, access, source_type, authors}], "unverified_numbers": [...], "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": {
      "topic": "<topic>"
    },
    "name": "deepseek_research"
  }
}

Input Schema

NameRequiredDescriptionDefault
langNoLanguage code such as 'en' or 'pt-BR' to restrict search sources to (default None)
depthNoDepth tier controlling queries, sources and reasoning (default 'medium')medium
topicYesComplex research topic or question to investigate (non-empty, max 500 chars)
refreshNoBypass local cache and perform fresh research pipeline (default False)
sourcesNoSource preference: 'any' (default), 'academic' (papers) or 'technical' (docs, specs, repos)any
synthesisNoWho writes the report: 'server' (default) or 'client' (you get ranked sources, write it, server verifies it)server
unverifiedNoSentences with numbers not found in their sources: 'flag' marks them (default), 'drop' removes themflag
strip_boilerplateNoRemove boilerplate from fetched source pages before synthesis (default True)

Output Schema

NameRequiredDescriptionDefault

No arguments

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed4 schema fields changedv3.3.0
    • changedInput schema / properties / lang / description
      Previous value: -"Two-letter ISO language code to restrict search sources to (default None)"New value: +"Language code such as 'en' or 'pt-BR' to restrict search sources to (default None)"
    • addedInput schema / properties / sources
      Added value: +{
      +  "default": "any",
      +  "description": "Source preference: 'any' (default), 'academic' (papers) or 'technical' (docs, specs, repos)",
      +  "enum": [
      +    "any",
      +    "academic",
      +    "technical"
      +  ],
      +  "title": "Sources",
      +  "type": "string"
      +}
    • addedInput schema / properties / synthesis
      Added value: +{
      +  "default": "server",
      +  "description": "Who writes the report: 'server' (default) or 'client' (you get ranked sources, write it, server verifies it)",
      +  "enum": [
      +    "server",
      +    "client"
      +  ],
      +  "title": "Synthesis",
      +  "type": "string"
      +}
    • addedInput schema / properties / unverified
      Added value: +{
      +  "default": "flag",
      +  "description": "Sentences with numbers not found in their sources: 'flag' marks them (default), 'drop' removes them",
      +  "enum": [
      +    "flag",
      +    "drop"
      +  ],
      +  "title": "Unverified",
      +  "type": "string"
      +}
  2. Changed16 schema fields changedv3.0.0
    • removedInput schema / properties / after
      Removed value: -{
      -  "anyOf": [
      -    {
      -      "type": "string"
      -    },
      -    {
      -      "type": "null"
      -    }
      -  ],
      -  "default": null,
      -  "title": "After"
      -}
    • removedInput schema / properties / before
      Removed value: -{
      -  "anyOf": [
      -    {
      -      "type": "string"
      -    },
      -    {
      -      "type": "null"
      -    }
      -  ],
      -  "default": null,
      -  "title": "Before"
      -}
    • removedInput schema / properties / dedupe
      Removed value: -{
      -  "default": true,
      -  "title": "Dedupe",
      -  "type": "boolean"
      -}
    • addedInput schema / properties / depth
      Added value: +{
      +  "default": "medium",
      +  "description": "Depth tier controlling queries, sources and reasoning (default 'medium')",
      +  "enum": [
      +    "low",
      +    "medium",
      +    "high",
      +    "max"
      +  ],
      +  "title": "Depth",
      +  "type": "string"
      +}
    • removedInput schema / properties / exclude_domains
      Removed value: -{
      -  "anyOf": [
      -    {
      -      "items": {
      -        "type": "string"
      -      },
      -      "type": "array"
      -    },
      -    {
      -      "type": "null"
      -    }
      -  ],
      -  "default": null,
      -  "title": "Exclude Domains"
      -}
    • removedInput schema / properties / include_domains
      Removed value: -{
      -  "anyOf": [
      -    {
      -      "items": {
      -        "type": "string"
      -      },
      -      "type": "array"
      -    },
      -    {
      -      "type": "null"
      -    }
      -  ],
      -  "default": null,
      -  "title": "Include Domains"
      -}
    • addedInput schema / properties / lang / description
      Added value: +"Two-letter ISO language code to restrict search sources to (default None)"
    • removedInput schema / properties / max_pages
      Removed value: -{
      -  "default": 5,
      -  "title": "Max Pages",
      -  "type": "integer"
      -}
    • removedInput schema / properties / per_page_chars
      Removed value: -{
      -  "default": 6000,
      -  "title": "Per Page Chars",
      -  "type": "integer"
      -}
    • removedInput schema / properties / question
      Removed value: -{
      -  "title": "Question",
      -  "type": "string"
      -}
    • addedInput schema / properties / refresh / description
      Added value: +"Bypass local cache and perform fresh research pipeline (default False)"
    • removedInput schema / properties / rewrite
      Removed value: -{
      -  "default": false,
      -  "title": "Rewrite",
      -  "type": "boolean"
      -}
    • addedInput schema / properties / strip_boilerplate / description
      Added value: +"Remove boilerplate from fetched source pages before synthesis (default True)"
    • addedInput schema / properties / topic
      Added value: +{
      +  "description": "Complex research topic or question to investigate (non-empty, max 500 chars)",
      +  "title": "Topic",
      +  "type": "string"
      +}
    • removedInput schema / properties / total_chars
      Removed value: -{
      -  "default": 30000,
      -  "title": "Total Chars",
      -  "type": "integer"
      -}
    • changedInput schema / required
      Previous value: -[
      -  "question"
      -]New value: +[
      +  "topic"
      +]
  3. Changed4 schema fields changedv2.0.1
    • addedInput schema / properties / after
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "After"
      +}
    • addedInput schema / properties / before
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Before"
      +}
    • addedInput schema / properties / exclude_domains
      Added value: +{
      +  "anyOf": [
      +    {
      +      "items": {
      +        "type": "string"
      +      },
      +      "type": "array"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Exclude Domains"
      +}
    • addedInput schema / properties / include_domains
      Added value: +{
      +  "anyOf": [
      +    {
      +      "items": {
      +        "type": "string"
      +      },
      +      "type": "array"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Include Domains"
      +}
  4. Changed4 schema fields changedv1.0.0
    • addedInput schema / properties / dedupe
      Added value: +{
      +  "default": true,
      +  "title": "Dedupe",
      +  "type": "boolean"
      +}
    • addedInput schema / properties / lang
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Lang"
      +}
    • addedInput schema / properties / rewrite
      Added value: +{
      +  "default": false,
      +  "title": "Rewrite",
      +  "type": "boolean"
      +}
    • addedInput schema / properties / strip_boilerplate
      Added value: +{
      +  "default": true,
      +  "title": "Strip Boilerplate",
      +  "type": "boolean"
      +}
  5. First observedv0.2.0

TDQS

A4.8/5.0
Behavior5/5

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

Annotations already declare readOnly/openWorld/idempotent, and the description still adds real behavioral context beyond them: cost scaling ($0.003 low to $0.03 max), a concurrency cap of 3 simultaneous jobs, the fact that sources are read but raw pages are not returned, citation verification behavior, [unverified] marking and how unverified="drop" alters output, and the client-synthesis verification path. That is exactly the kind of operational context annotations cannot carry.

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 purpose, then behavior, then routing, then return shape — good ordering with zero wasted preamble. It is dense and parenthetical-heavy, and the explicit 'Returns: {...}' block largely restates an output schema that already exists, which is the main redundancy.

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?

For an 8-parameter, costly, potentially asynchronous, open-world tool, the description covers the things an agent must know: cost, rate/concurrency limits, async continuation, citation verification semantics, and when to delegate to a sibling. Nothing needed to invoke it correctly appears missing.

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; the description earns above baseline by mapping each depth tier to a use case, spelling out what 'academic' and 'technical' source preferences actually select, and explaining the observable output effect of unverified flag vs drop and of synthesis="client". lang, refresh and strip_boilerplate are left entirely to the schema, which is acceptable given full coverage but leaves the addition partial.

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?

Opens with a specific verb and deliverable: research a topic across the web and return a cited report in one call. It also implicitly distinguishes itself from the sibling search/fetch/extract tools by naming what those do — 'one fact or query (deepseek_search)', 'known URLs (deepseek_fetch)' — so an agent can tell deepseek_research apart without inspecting schemas.

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

Usage Guidelines5/5

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

Explicit 'When NOT to use' block routes to deepseek_search, deepseek_search_many, deepseek_fetch and deepseek_extract with the exact condition that selects each. It also gives positive selection guidance per depth tier ('high' for decisions with several viewpoints, 'max' for broad or contested topics) and an async rule ('If status is "running", call deepseek_check_research with the job_id; do not start the same research again').

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