Skip to main content
Glama

Web Search

infobroker_search_web
Read-onlyIdempotent

Search the web, encyclopedia, academic, and code sources through one interface with automatic provider selection and a fallback chain.

Instructions

Search the web, encyclopedia, academic, and code sources through one interface with automatic provider selection and a fallback chain. Use when you need broad or batched search, query autocomplete (suggest), query expansion (expand), ranked passages from the top pages (deep), or a multi-variant research compile that deep-reads each variant (research), and optional in-process analysis of the returned results (analyze: topics, entities, clusters, timeline). Do NOT use for a URL you already have (use infobroker_fetch_page), for high-stakes claim verification (use infobroker_verify_claims), for stored-report recall (use infobroker_manage_kb search), or for academic citations (use infobroker_get_citations). Caches results in the local knowledge base, enforces per-provider rate limits, and needs no API key for the free providers. task_type forces a dispatch chain (for example academic) when the automatic classification picks the wrong one. fresh: false keeps KB-first sufficiency, fresh: true skips KB recall and searches external providers, and fresh: 'blend' keeps cached KB results alongside fresh external results. collection scopes KB recall to a named collection. audience records who will read the result and is echoed in meta.direction with the executed task type, reading depth, result bound and provider pool. When two task types are equally supported and would reach different providers, the tool returns both candidate requirements with an empty result list (meta.direction.ambiguous) instead of searching, so the caller can re-issue with task_type. include_domains restricts results to or prefers listed domains (include_domains_mode: restrict filters, prefer ranks them ahead), on top of the configured blocklist; the response reports the matched and excluded counts. An array query is searched in parallel and merged, and a failed element is reported in meta.per_query without failing the batch. A measurable attribute (attendance, population, area, elevation, percentage, date) returns a sourced facts entry before prose, with meta.fact_status. Returns a JSON envelope prefixed [OK] or [ERROR] with status, provider, results, and meta.

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": {
      "query": "<query>"
    },
    "name": "infobroker_search_web"
  }
}

Input Schema

NameRequiredDescriptionDefault
deepNoRead the top results and return each page's ranked passages against the query
pageNoResults page number (default 1)
freshNoKnowledge-base recall mode: false keeps KB-first sufficiency, true skips recall and searches external providers, blend keeps KB results alongside fresh external results (default false)
queryYesSearch query: a single string, or up to five strings searched in parallel
expandNoReturn query-expansion strings instead of search results
regionNoISO region/country code (e.g. 'us-en', 'DE')
analyzeNoAttach in-process analysis to the response: topic terms, entity mentions, semantic result clusters, and a dated timeline. Derived from the returned results only; adds no outbound request (default false)
suggestNoReturn query-autocomplete strings instead of results (default false)
audienceNoWho will read the result: executive, analyst, engineer, policy, legal, journalist, or investor. Recorded in the direction block and carried into a report manifest; it calibrates depth and emphasis, never facts
languageNoPreferred result language (ISO 639-1, e.g. 'de'). Providers that cannot honor it report it in meta.ignored_params
priorityNoRoute by intent: speed, quality, privacy, or free_only
providerNoProvider slug (auto-select if omitted)
researchNoResearch compile: fan out to derived query variants and deep-read each, returning passages grouped by variant (default off)
task_typeNoForce a dispatch chain by task type (auto-classified when omitted)
collectionNoKnowledge-base collection to recall from (scopes KB recall only; ignored when fresh is true)
time_rangeNoRestrict results to day, week, month, or year
max_resultsNoNumber of results to return (1-30, default 8)
safe_searchNoSafe-search filtering: on, off, or strict (default on)on
content_typeNoSource kind to search: docs, issue, changelog, blog, spec, or all (default all)all
include_domainsNoDomains to restrict or prefer (max 300). Matching is case-insensitive and a domain matches its subdomains; an empty list is a no-op
include_domains_modeNoHow include_domains applies: restrict returns only matching hosts, prefer ranks them ahead without excluding any (default restrict)restrict

Output Schema

NameRequiredDescriptionDefault
metaNo
errorNo
statusYes
noticesNo
resultsNo
providerYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changedv2026.11.16
    • addedInput schema / properties / audience
      Added value: +{
      +  "description": "Who will read the result: executive, analyst, engineer, policy, legal, journalist, or investor. Recorded in the direction block and carried into a report manifest; it calibrates depth and emphasis, never facts",
      +  "enum": [
      +    "executive",
      +    "analyst",
      +    "engineer",
      +    "policy",
      +    "legal",
      +    "journalist",
      +    "investor"
      +  ],
      +  "type": "string"
      +}
    • addedInput schema / properties / include_domains
      Added value: +{
      +  "description": "Domains to restrict or prefer (max 300). Matching is case-insensitive and a domain matches its subdomains; an empty list is a no-op",
      +  "items": {
      +    "type": "string"
      +  },
      +  "maxItems": 300,
      +  "type": "array"
      +}
    • addedInput schema / properties / include_domains_mode
      Added value: +{
      +  "default": "restrict",
      +  "description": "How include_domains applies: restrict returns only matching hosts, prefer ranks them ahead without excluding any (default restrict)",
      +  "enum": [
      +    "restrict",
      +    "prefer"
      +  ],
      +  "type": "string"
      +}
  2. Changed7 schema fields changedv2026.10.30
    • addedInput schema / properties / collection
      Added value: +{
      +  "description": "Knowledge-base collection to recall from (scopes KB recall only; ignored when fresh is true)",
      +  "type": "string"
      +}
    • addedInput schema / properties / fresh / anyOf
      Added value: +[
      +  {
      +    "type": "boolean"
      +  },
      +  {
      +    "const": "blend",
      +    "type": "string"
      +  }
      +]
    • changedInput schema / properties / fresh / description
      Previous value: -"Skip the knowledge-base recall shortcut and search external providers (default false)"New value: +"Knowledge-base recall mode: false keeps KB-first sufficiency, true skips recall and searches external providers, blend keeps KB results alongside fresh external results (default false)"
    • removedInput schema / properties / fresh / type
      Removed value: -"boolean"
    • changedInput schema / properties / query / anyOf
      Previous value: -[
      -  {
      -    "description": "Search query",
      -    "type": "string"
      -  },
      -  {
      -    "description": "Multiple queries to search in parallel (max 5)",
      -    "items": {
      -      "type": "string"
      -    },
      -    "maxItems": 5,
      -    "type": "array"
      -  }
      -]New value: +[
      +  {
      +    "description": "Search query",
      +    "type": "string"
      +  },
      +  {
      +    "description": "Multiple queries to search in parallel (max 5); a failed element is reported per input without failing the batch",
      +    "items": {
      +      "type": "string"
      +    },
      +    "maxItems": 5,
      +    "type": "array"
      +  }
      +]
    • addedInput schema / properties / task_type
      Added value: +{
      +  "description": "Force a dispatch chain by task type (auto-classified when omitted)",
      +  "enum": [
      +    "general_web",
      +    "small_web",
      +    "encyclopedia",
      +    "definition",
      +    "structured_fact",
      +    "financial",
      +    "location",
      +    "academic",
      +    "code",
      +    "news",
      +    "archive",
      +    "semantic",
      +    "synthesis",
      +    "privacy_critical"
      +  ],
      +  "type": "string"
      +}
    • changedOutput schema / (root)
      Previous value: -nullNew value: +{
      +  "$schema": "http://json-schema.org/draft-07/schema#",
      +  "additionalProperties": {},
      +  "properties": {
      +    "error": {
      +      "additionalProperties": {},
      +      "properties": {
      +        "code": {
      +          "type": "string"
      +        },
      +        "message": {
      +          "type": "string"
      +        },
      +        "remediation": {
      +          "type": "string"
      +        }
      +      },
      +      "required": [
      +        "code",
      +        "message"
      +      ],
      +      "type": "object"
      +    },
      +    "meta": {
      +      "additionalProperties": {},
      +      "properties": {},
      +      "type": "object"
      +    },
      +    "notices": {
      +      "items": {
      +        "additionalProperties": {},
      +        "properties": {},
      +        "type": "object"
      +      },
      +      "type": "array"
      +    },
      +    "provider": {
      +      "type": "string"
      +    },
      +    "results": {
      +      "items": {
      +        "additionalProperties": {},
      +        "properties": {},
      +        "type": "object"
      +      },
      +      "type": "array"
      +    },
      +    "status": {
      +      "enum": [
      +        "ok",
      +        "error"
      +      ],
      +      "type": "string"
      +    }
      +  },
      +  "required": [
      +    "status",
      +    "provider"
      +  ],
      +  "type": "object"
      +}
  3. Changed2 schema fields changed
    • addedInput schema / properties / analyze
      Added value: +{
      +  "default": false,
      +  "description": "Attach in-process analysis to the response: topic terms, entity mentions, semantic result clusters, and a dated timeline. Derived from the returned results only; adds no outbound request (default false)",
      +  "type": "boolean"
      +}
    • addedInput schema / properties / language
      Added value: +{
      +  "description": "Preferred result language (ISO 639-1, e.g. 'de'). Providers that cannot honor it report it in meta.ignored_params",
      +  "type": "string"
      +}
  4. Changed1 schema field changedv2026.10.22
    • addedInput schema / properties / fresh
      Added value: +{
      +  "default": false,
      +  "description": "Skip the knowledge-base recall shortcut and search external providers (default false)",
      +  "type": "boolean"
      +}
  5. Addedv0.1.2

TDQS

A4.6/5.0
Behavior4/5

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

Annotations already declare read-only, open-world, idempotent, non-destructive, so the safety bar is lower; the description still adds substantial non-obvious behavior: local KB caching, per-provider rate limits, no API key needed on free providers, per-query failure isolation in meta.per_query, and the unusual ambiguity path that returns two candidate requirement sets with an empty result list instead of searching. The only gap is that the [OK]/[ERROR] envelope contents are sketched rather than detailed, which is partly covered by the output schema.

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 correctly: purpose sentence first, then when-to-use, then when-not-to-use, then behavioral notes. It is dense and runs long for one paragraph, and several clauses (fact_status, per_query, ambiguity) are packed together, but nearly every sentence carries routing or behavior information rather than filler.

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 a 21-parameter tool with 7 enums, an output schema, and full annotations, the description covers the decision-critical dimensions: routing, caching/freshness semantics, domain filtering, batching failure behavior, ambiguity handling, and fact extraction. Combined with the 100%-covered schema and existing output schema, an agent has enough to select and invoke it correctly.

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 the baseline is 3, but the description genuinely adds meaning on top: it explains that task_type forces a dispatch chain when auto-classification misfires, spells out all three fresh modes (KB-first, external-only, blend), notes that collection is ignored when fresh is true, says audience is echoed in meta.direction, and explains restrict-vs-prefer interaction with the configured blocklist plus matched/excluded counts. That is real semantic value beyond the field descriptions.

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?

States a specific verb (search) and resource set (web, encyclopedia, academic, code) plus the mechanism (single interface, automatic provider selection, fallback chain). It explicitly separates itself from infobroker_fetch_page, verify_claims, manage_kb, and get_citations, so an agent can route without opening another schema.

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?

Both sides of the routing decision are spelled out: a positive 'Use when...' list (broad/batched search, suggest, expand, deep, research, analyze) and an explicit 'Do NOT use for...' list naming the correct sibling for each excluded case (existing URL, high-stakes verification, stored-report recall, citations). That is the full when/when-not/alternative triad.

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