Skip to main content
Glama
merterbak

Grok MCP

by merterbak
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python

web_search

Read-only

Answer research questions by searching the live web across multiple turns, optionally analyzing images, and returning a synthesized response with cited sources.

Instructions

Answer a query using agentic real-time web search.

Grok browses the web across multiple turns, optionally inspecting images on
pages, then synthesizes an answer with citations.

Args:
    prompt: Search query or research question.
    model: Grok model used to drive the agent (default `grok-4.6`).
    allowed_domains: Restrict search to these domains (max 5, mutually exclusive with excluded).
    excluded_domains: Exclude these domains from search (max 5).
    enable_image_understanding: Let the agent analyze images it encounters.
    enable_image_search: Let the agent search for and return image results.
    include_inline_citations: Embed `[1]`-style citation markers into the answer text.
    max_turns: Cap the agent's reasoning/tool turns.
    show_usage: Append a token usage and cost footer to the answer (default False).

Returns:
    Markdown with the answer body followed by a `**Sources:**` list of cited URLs.

Input Schema

NameRequiredDescriptionDefault
modelNogrok-4.6
promptYes
max_turnsNo
show_usageNo
allowed_domainsNo
excluded_domainsNo
enable_image_searchNo
include_inline_citationsNo
enable_image_understandingNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv1.0.4
  2. Removed
  3. Changed1 schema field changedv1.0.1
    • addedInput schema / properties / show_usage
      Added value: +{
+  "default": false,
+  "title": "Show Usage",
+  "type": "boolean"
+}
  4. Changed1 schema field changed
    • addedInput schema / properties / enable_image_search
      Added value: +{
+  "default": false,
+  "title": "Enable Image Search",
+  "type": "boolean"
+}
  5. Added
  6. Removed
  7. Added
  8. Removed
  9. Added
  10. Removed
  11. Changed1 schema field changed
    • changedInput schema / properties / model / default
      Previous value: -"grok-4-1-fast"New value: +"grok-4-1-fast-reasoning"
  12. Added

TDQS

A4.2/5.0
Behavior4/5

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

Annotations only declare readOnlyHint=true; the description goes well beyond that by disclosing the multi-turn agentic loop, optional image inspection/search, citation embedding, and the exact return shape (Markdown body plus a '**Sources:**' URL list). It adds real behavioral context, though it does not discuss latency, cost, or rate behavior beyond the show_usage footer.

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

Conciseness5/5

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

The document is front-loaded with a one-line answer-in-itself summary, then a compact behavior paragraph, then an Args/Returns block where every line earns its place. Nothing is redundant despite covering 9 parameters.

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 no output schema, the description compensates by specifying the return format (Markdown answer plus Sources list), and it documents every input plus the agentic search behavior. An agent has everything needed to call this correctly without opening anything else.

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

Parameters5/5

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

Schema description coverage is 0% with 9 parameters, so the description carries the full burden and does so: it explains max-5 caps and mutual exclusivity for allowed/excluded_domains, the model default, what max_turns caps, and what enable_image_understanding/search and include_inline_citations actually do. This is meaningfully more than the bare parameter names.

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

Purpose4/5

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

The opening sentence gives a specific verb+resource ('Answer a query using agentic real-time web search') and the follow-up clarifies the mechanism (multi-turn browsing, image inspection, cited synthesis). It does not, however, name or contrast itself with the sibling x_search, so an agent must infer the boundary between web and X search.

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

Usage Guidelines3/5

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

Usage context is implied by the purpose statement – use it when you need a cited, web-grounded answer – but there is no explicit when-to-use/when-not guidance and no reference to alternatives like x_search or chat. The domain-restriction flags hint at scoping use cases, but selection guidance is left to inference.

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

Deploy Server

Other Tools