SuanLinFang AI Capabilities

Recognize Image

recognize_image

识别图片中的文字（png/jpg/webp/bmp，≤20MB，公网 URL），同步返回逐行文本。 tier：standard 标准档（快速，适合清晰打印件/截图）；accurate 精准档（表格、多栏、倾斜拍摄等复杂版面）。 prompt 仅精准档生效：自定义识别指令（如「只提取表格金额列」），≤2000 字符，缺省识别全部文字。计费按页（单图=1 页，约 0.5~1.5 分/页）。

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": {
      "image_url": "<image_url>",
      "tier": "<tier>"
    },
    "name": "recognize_image"
  }
}

Input Schema

Name	Required	Description	Default
`tier`	Yes
`prompt`	No
`image_url`	Yes

Output Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

First observedNov 11, 2026

TDQS

A4.2/5.0

Behavior4/5

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

No annotations exist, so the description carries the full burden, and it does well: it discloses sync (not task-based) return, the 20MB cap, the public-URL requirement, per-page billing (~0.5–1.5 credits/page, single image = 1 page), and the tier-scoped effect of prompt. Not covered: failure behavior, language coverage, or whether the URL must be reachable without auth.

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?

The content is dense but front-loaded, with the core action first, followed by labeled clauses for tier, prompt, and pricing. Each sentence carries distinct information with no filler. The compact multi-clause layout is efficient but slightly more terse than ideal for scanning.

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?

For a three-parameter synchronous OCR tool with an output schema (so return structure needn't be described), the definition covers input constraints, tier semantics, prompt rules, and cost. Remaining gaps are minor: no statement of language support or error/edge-case handling.

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%, so the description must explain every parameter itself, and it does. image_url gets format, size, and public-URL constraints; tier's two enum values are defined with use cases; prompt's tier gating, 2000-character limit, default of null (recognize all text), and example usage are all specified. Nothing about the three parameters is left ambiguous.

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 description states a specific verb and resource: it recognizes text in an image (OCR) and returns line-by-line text synchronously. It also pins down the accepted formats (png/jpg/webp/bmp) and size limit (≤20MB). It never names a sibling, so an agent must infer that documents/audio go elsewhere, which keeps it just short of a 5.

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?

Tier selection is explained with concrete conditions: standard for clean printouts and screenshots, accurate for tables, multi-column layouts, and tilted photos. It also states that prompt only takes effect in the accurate tier. What's missing is any guidance on when to use this synchronous tool versus an async submit_* sibling, or how images differ from documents.

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

All SuanLinFang AI Capabilities tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?