Skip to main content
Glama
Sign In

Keryx

by io.github.tang-vu
OverviewDiscussionsAdmin

Check a draft's cited claims against supplied source text

draft_check
Read-onlyIdempotent

Split a private passage into sentences, bind citation markers ([3], cite{key}, [@key], Author (Year)) to a BibTeX, RIS or DOI/link reference list, and compare each cited sentence with source text you supply. Each verdict (supported, partly-supported, not-found, source-unavailable) carries a verbatim excerpt of that text or states that no supporting passage was found; a source without supplied text is always source-unavailable. Deterministic wording comparison, not semantic review. No fetching, model, database, history, payment or reward I/O. Inputs and results can be retained by your MCP client; choose that client’s privacy settings before sharing private passages.

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": "draft_check"
  }
}

Input Schema

NameRequiredDescriptionDefault
checkNoVersion-1 private-draft-check contract, method lexical-overlap-v2, at most 65536 UTF-8 bytes. Optional sources[].identity binds exact DOI/URL and contentVersion; suggestions[] stays separate from written citations. No fetching or payment. See docs/draft-check.md.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / check / description
      Previous value: -"Version-1 private-draft-check contract, at most 65536 UTF-8 bytes. See docs/draft-check.md."New value: +"Version-1 private-draft-check contract, method lexical-overlap-v2, at most 65536 UTF-8 bytes. Optional sources[].identity binds exact DOI/URL and contentVersion; suggestions[] stays separate from written citations. No fetching or payment. See docs/draft-check.md."
  2. Added

TDQS

A4.6/5.0
Behavior5/5

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

Annotations only declare read-only/idempotent/closed-world, but the description adds real behavioral context: deterministic lexical comparison rather than semantic review, the four verdict types with verbatim excerpts, the rule that a source without supplied text is always source-unavailable, the absence of network/model/database/history/payment I/O, and a client-side retention/privacy warning.

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?

Dense and front-loaded: the core split-bind-compare action leads, verdict/return behavior follows, and constraints and the privacy caveat are correctly placed last. Five information-packed sentences are on the long side, but each earns its place and nothing is redundant.

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

Completeness5/5

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

With no output schema, the description carries the burden of explaining return values and does so (verdict types plus verbatim excerpts), while annotations cover the safety profile. For a one-parameter tool with closed-world, idempotent, read-only behavior, nothing needed to call it correctly is missing.

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?

The single parameter is fully documented in the schema (100% coverage), which sets a baseline of 3, but the description adds format-level meaning the schema lacks: the citation markers accepted ([3], cite{key}, [@key], Author (Year)) and the reference-list formats (BibTeX, RIS, DOI/link). That goes beyond restating 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 chain (split, bind, compare) over a precise resource (a private passage's cited claims vs supplied source text) and enumerates the exact verdict vocabulary. It is clearly distinguishable from siblings like evidence_draft or paper_lookup without opening any 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 clear context for when this applies (comparing a draft's cited claims against source text the caller supplies) and implicitly rules out fetching by stating 'No fetching, model, database, history...'. It never names an alternative sibling or an explicit when-not condition, so it stops 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.

All Keryx toolsInspect
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