Skip to main content
Glama
octen-mcp

octen-mcp

Official
by Octen-Team

Image Search

image_search
Read-only

Find images on the web by text query or reference image and get ranked results with titles, sources, dimensions, thumbnails, and summaries. Use for photos, diagrams, screenshots, or visual references.

Instructions

Find images on the web by text query OR by a reference image — returns ranked results (title, source page, dimensions, thumbnail, description, summary). Pass exactly one of: a text query, an image_url (a picture already on the web), or image_data (base64, for a picture you hold). Never more than one. Use this when the user wants pictures, photos, diagrams, screenshots, or visual references — not for general text web search.

keywords: find images, image search, photos, pictures, screenshots, visual reference, diagram, icon, illustration, reference image

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "image_search"
  }
}

Input Schema

NameRequiredDescriptionDefault
countNoNumber of results to return (1-10). Default 5.
queryNoText query describing the images to find. Exactly one of `query`, `image_url` or `image_data` — never more than one.
timeoutNoRequest timeout in seconds (1-60). Defaults to 30s if unset.
image_urlNoPublic image URL to search by visual similarity. Exactly one of `query`, `image_url` or `image_data` — never more than one.
image_dataNoBase64-encoded image to search by visual similarity, for an image you hold rather than one already on the web. At most 5MB encoded; JPEG, PNG, WEBP, BMP, TIFF, ICO, DIB, ICNS or SGI. A `data:` URI is accepted — its payload is used. Exactly one of `query`, `image_url` or `image_data` — never more than one.
safesearchNoAdult-content filter. Default strict.strict
exclude_domainsNoDrop results from these domains.
include_domainsNoOnly return results from these domains (e.g. 'unsplash.com').

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changedv0.7.1
    • removedInput schema / properties / html_snippet
      Removed value: -{
      -  "description": "Return an HTML snippet of the source context per result. Omit to use the server default.",
      -  "properties": {
      -    "enable": {
      -      "description": "Whether to return HTML snippets.",
      -      "type": "boolean"
      -    },
      -    "max_tokens": {
      -      "default": 5000,
      -      "description": "Max tokens per HTML snippet (100-100000). Default 5000.",
      -      "maximum": 100000,
      -      "minimum": 100,
      -      "type": "integer"
      -    }
      -  },
      -  "type": "object"
      -}
    • removedInput schema / properties / topic
      Removed value: -{
      -  "default": "general",
      -  "description": "Image category: `general` for broad image search, `design` for design / illustration oriented results. Default general.",
      -  "enum": [
      -    "general",
      -    "design"
      -  ],
      -  "type": "string"
      -}
  2. Changed9 schema fields changedv0.5.0
    • removedInput schema / properties / end_time
      Removed value: -{
      -  "description": "Upper bound for the time window, ISO 8601.",
      -  "type": "string"
      -}
    • changedInput schema / properties / html_snippet / properties / max_tokens / description
      Previous value: -"Max tokens per HTML snippet (min 100). Default 5000."New value: +"Max tokens per HTML snippet (100-100000). Default 5000."
    • addedInput schema / properties / html_snippet / properties / max_tokens / maximum
      Added value: +100000
    • addedInput schema / properties / image_data
      Added value: +{
      +  "description": "Base64-encoded image to search by visual similarity, for an image you hold rather than one already on the web. At most 5MB encoded; JPEG, PNG, WEBP, BMP, TIFF, ICO, DIB, ICNS or SGI. A `data:` URI is accepted — its payload is used. Exactly one of `query`, `image_url` or `image_data` — never more than one.",
      +  "maxLength": 5242880,
      +  "type": "string"
      +}
    • changedInput schema / properties / image_url / description
      Previous value: -"Optional public image URL to search by reference image (visual similarity), in addition to the text `query`."New value: +"Public image URL to search by visual similarity. Exactly one of `query`, `image_url` or `image_data` — never more than one."
    • changedInput schema / properties / query / description
      Previous value: -"Text query describing the images to find."New value: +"Text query describing the images to find. Exactly one of `query`, `image_url` or `image_data` — never more than one."
    • removedInput schema / properties / start_time
      Removed value: -{
      -  "description": "Lower bound for the time window, ISO 8601 (e.g. '2025-01-01T00:00:00Z').",
      -  "type": "string"
      -}
    • removedInput schema / properties / time_range
      Removed value: -{
      -  "description": "Relative time window (e.g. `week`, `month`). Mutually exclusive with `start_time`/`end_time` — if both are given, the absolute range wins.",
      -  "enum": [
      -    "day",
      -    "week",
      -    "month",
      -    "year",
      -    "d",
      -    "w",
      -    "m",
      -    "y"
      -  ],
      -  "type": "string"
      -}
    • removedInput schema / required
      Removed value: -[
      -  "query"
      -]
  3. Changed1 schema field changedv0.4.2
    • changedInput schema / properties / timeout / description
      Previous value: -"Request timeout in seconds (1-60)."New value: +"Request timeout in seconds (1-60). Defaults to 30s if unset."
  4. Addedv0.3.1

TDQS

A4.3/5.0
Behavior4/5

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

Annotations already declare readOnlyHint=true, openWorldHint=true and destructiveHint=false, so safety is covered structurally. The description adds real value beyond that: it enumerates the returned fields (title, source page, dimensions, thumbnail, description, summary) and states the hard mutual-exclusion rule on the three input modes. Rate limits/quotas and failure behavior are still unstated.

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?

Two tight sentences, front-loaded with the core capability, then the mode constraint, then the disambiguation. The trailing keyword list is somewhat redundant with the prose but is cheap and aids retrieval; nothing else is wasted.

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

Completeness4/5

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

With eight parameters and no output schema, the description carries appropriate weight: it covers mode selection and describes the return shape. Gaps are minor — nothing about pagination, result ranking behavior, or when the domain filters are worth using — but an agent could call this correctly as written.

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; the description earns above that by framing the parameters as three mutually exclusive search modes and explaining the intent behind image_data ("a picture you hold") versus image_url ("a picture already on the web"). count, timeout, safesearch and the domain filters are left entirely to the schema.

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+resource ("Find images on the web") and enumerates the three input modes (text query, image_url, image_data) plus the result shape. It also draws a boundary against text web search, so an agent can distinguish it from `search` and `broad_search` without opening a schema.

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?

Gives an explicit positive trigger ("pictures, photos, diagrams, screenshots, or visual references") and an explicit negative one ("not for general text web search"). It stops short of naming the adjacent sibling tools by identifier (e.g. `search`, `video_search`), so the agent must infer which sibling to fall back to for video or text.

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