Skip to main content
Glama

verify_citations

Checks all [[id]] citations in a text against the wiki and returns valid and invalid IDs, detecting hallucinated references after an LLM response.

Instructions

주어진 텍스트의 모든 [[id]] 인용을 wiki와 대조 — hallucination 검출.

LLM 응답 직후 호출하면 존재하지 않는 ID를 인용했는지 확인 가능.

Args: text: 검증할 텍스트 (LLM 응답)

Returns: 총 인용 수, 유효, 무효 ID 목록

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": {
      "text": "<text>"
    },
    "name": "verify_citations"
  }
}

Input Schema

NameRequiredDescriptionDefault
textYes

Output Schema

NameRequiredDescriptionDefault
resultYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.3.0

TDQS

A4.3/5.0
Behavior4/5

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

With no annotations, the description carries the burden and does disclose the core behavior: it compares citations against the wiki rather than trusting the text, and it reports total/valid/invalid counts. It never states explicitly that the operation is side-effect free or what wiki scope is consulted, but the verification framing makes mutation implausible.

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 purpose line, then a usage note, then Args/Returns blocks — efficient and scannable. The Returns block partially duplicates the existing output schema, which is minor waste but costs little.

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 one-parameter read-style tool with an output schema, the definition covers purpose, triggering context, input meaning, and output shape. The only gap is which wiki corpus or domains the comparison spans, which an agent might want before trusting the result.

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 description coverage is 0%, so the description must compensate, and it does: the single `text` parameter is characterized as the text to verify, specifically an LLM response containing `[[id]]` citations. That adds real meaning beyond the bare `string` type in 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 and resource: cross-check every `[[id]]` citation in the given text against the wiki to detect hallucinated IDs. The concrete citation syntax `[[id]]` plus the hallucination-detection framing makes it distinguishable from siblings like find_broken_links, which operates on wiki links rather than citations inside an arbitrary text blob.

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?

Explicitly says to call it immediately after an LLM response to check whether non-existent IDs were cited, which gives a clear triggering context. It stops short of naming alternatives or stating when not to use it, so it is short of a full 5.

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