Skip to main content
Glama

Check facts in a draft against the records

check_claims
Read-onlyIdempotent

Check the facts of a draft (a capture brief, a market report, an e-mail) against the records before it goes out. Send each fact as a claim; each comes back matches, differs (with what the record says), not_found or not_checked, and the page that shows it. A dollar figure matches when the record rounds to the words it was stated in ("$4.2 billion" covers 4.15 to 4.25). Nothing is judged by an AI here: every claim is a lookup. Use it after writing, on the figures, dates, names, certifications and quoted rules the draft relies on; fix or drop what differs, and say what could not be checked.

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": {
      "claims": "<claims>"
    },
    "name": "check_claims"
  }
}

Input Schema

NameRequiredDescriptionDefault
claimsYesThe facts to check, up to 25, each an object with `kind` and, optionally, `text` (the sentence it comes from). amount: {"kind": "amount", "plan": <QueryPlan>, "stated": "$4.2 billion"} (or "value": 4200000000; "row": "Leidos" picks a row of a ranking). notice: {"kind": "notice", "notice_id": "<id>", "deadline": "2026-12-01", "set_aside": "SDVOSB", "naics": "541512"}. contract: {"kind": "contract", "award": "<award key or number>", "incumbent": "Leidos", "end_date": "2027-03-31", "stated": "$12.5M"}. company: {"kind": "company", "company": "<UEI or name>", "certification": "8(a)", "state": "VA", "registered": true, "excluded": false}. far: {"kind": "far", "section": "19.502-2", "quote": "<words quoted>"}.

Output Schema

NameRequiredDescriptionDefault

No arguments

Schema Changelog

Changes observed during successful MCP inspections.

  1. Added

TDQS

A4.1/5.0
Behavior4/5

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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive, closed-world), so the description's added value is in semantics: it declares that no AI judgment is involved, that every claim is a lookup, and gives a concrete tolerance rule ('$4.2 billion' covers 4.15 to 4.25). What it does not disclose is behavior on malformed or unknown claim kinds, or what happens past the claim limit, which is why this is a 4 rather than a 5.

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 the action, then result codes, then the matching rule, then usage — a sensible order with no filler. It is slightly dense in a single paragraph and the usage instruction is tacked onto the end rather than separated, but every sentence carries information an agent needs.

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 read-only batch verification tool with an output schema already covering return shape, the description supplies the missing pieces: what a claim is, how matching is decided, and what to do with 'differs' and 'not_checked' results. Error handling for unsupported claim kinds is the main gap, but that is minor for a 1-parameter, fully documented tool.

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

Parameters3/5

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

Schema description coverage is 100% and the schema already enumerates each claim kind with literal example objects, so the description is at baseline. 'Send each fact as a claim' adds the conceptual framing and the rounding rule clarifies 'stated' versus 'value', but nothing about kind-specific fields or validation behavior is added beyond 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 — check the facts of a draft (capture brief, market report, e-mail) against the records — and names the exact result taxonomy (matches, differs, not_found, not_checked). It also explicitly positions itself as a deterministic lookup rather than an AI judgment, which separates it from the search/lookup siblings an agent might otherwise pick.

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 clear when-to-use guidance: 'Use it after writing, on the figures, dates, names, certifications and quoted rules the draft relies on', followed by the follow-up action ('fix or drop what differs'). It does not state when NOT to use it or name an alternative such as get_notice or company_profile for an agent that only wants a single record, so it falls short of full routing guidance.

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.

Resources