Skip to main content
Glama

Найти персональные данные

find_personal_data

Находит персональные данные в PDF, файл не меняет. Номера проверяются контрольными суммами (СНИЛС, ИНН, карты). file_url: публичный URL PDF-файла. kinds: passport, snils, inn, phone, email, card, account; по умолчанию всё. words: свои слова — ФИО, адрес (каждое от 3 символов). Возвращает result: {"pages": [{"page", "items": [{"kind", "text", "boxes"}]}], "scanned": [страницы-сканы без текста], "total"}; boxes — [x0, y0, x1, y1] долями страницы, их можно передать в redact_pdf как areas.

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": {
      "file_url": "<file_url>"
    },
    "name": "find_personal_data"
  }
}

Input Schema

NameRequiredDescriptionDefault
kindsNo
wordsNo
file_urlYes

Output Schema

NameRequiredDescriptionDefault

No arguments

Schema Changelog

Changes observed during successful MCP inspections.

  1. Added

TDQS

A4.4/5.0
Behavior4/5

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

Adds behavior annotations don't carry: the file is not modified, numeric matches are validated by checksums (СНИЛС, ИНН, cards) which suppresses false positives, and "scanned" reports pages lacking a text layer so the agent knows detection is incomplete there. Note the mild tension with readOnlyHint=false, but destructiveHint=false is consistent with the no-change claim (non-read-only likely reflects a paid remote fetch, hence openWorldHint=true).

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 and the no-modification guarantee, then compact per-parameter lines and a return-structure line. Dense and waste-free, though the squeezed param/return notation is closer to notes than prose and takes a beat longer to parse.

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?

A rich output schema exists, yet the description still explains the meaningful parts — pages/items/boxes, the "scanned" list, and that boxes are fractional page coordinates reusable as areas in redact_pdf. Missing only operational limits (file size, page caps, whether scanned PDFs need prior OCR), which matter for a remote open-world fetch.

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 coverage is 0%, so the description must carry all meaning — and it does: file_url must be a *public* URL to a PDF, kinds enumerates every accepted value with the default of "всё", and words defines custom strings (ФИО, адрес) with a minimum length of 3 characters. Every one of the 3 parameters is documented beyond the bare types.

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+resource: "Находит персональные данные в PDF, файл не меняет" — a detection/scan operation, not an edit. It immediately separates itself from mutation siblings (redact_pdf, edit_pdf) and from generic text extractors by naming the PII kinds it targets.

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?

Clear context for use (scanning a public PDF for PII) and an explicit downstream hand-off: boxes "можно передать в redact_pdf как areas" — that is a real routing signal to a sibling. It stops short of explicit exclusions (e.g., when to prefer ocr_pdf or pdf_to_txt for text-layer-less files), but the pipeline guidance is genuinely useful.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.