Independent second opinion on any AI answer: inaccuracies, gaps, and a scored report card.
Scored across 1 tool
There is only one tool, so there is nothing within the set to confuse it with. Its scope is drawn tightly with an explicit 'Not for:' list that pushes adjacent use cases (fact-checking, document Q&A, AI-text detection, editing) away, which further reduces misselection risk.
The single name, submit_review_request, follows a clean verb_noun pattern that reads naturally and predicts the action. No competing conventions exist to clash with it.
One tool is a thin surface even for a focused review service; the rubric treats 1-2 tools as borderline. A single heavyweight entry point is defensible given the narrow purpose, but it leaves no room for status, history, or follow-up operations.
Submission is covered, and the description implies the verdict is returned inline, but there is no way to retrieve a past review, check status on a long-running panel, or ask follow-up questions about a returned verdict. That last gap is notable because the tool advertises follow-up questions as an output yet explicitly excludes discussing them, creating a dead end.
Independent review of AI-generated work in this conversation: your answer, or code, a pull request, a plan, an analysis or a draft produced with AI. Covers requests such as: is this right, a hallucination check, a fact-check, what's missing, challenge my assumptions, an adversarial review, a gate review, or an independent sign-off before merging, shipping or submitting. A panel of reviewers running on another company's AI model returns a verdict, inaccuracies with corrections, gaps, follow-up questions and a scored report card.
The review covers the complete text submitted, so an unabridged copy of the work gives a complete result; a revised draft can be reviewed again.
Not for: verifying a single factual claim against sources, questions about an uploaded document, detecting whether text was AI-written, evaluating the user's own AI agents, writing or editing, or discussing a review that has already been returned ("Do you agree?", "Was that fair?").
| Name | Required | Description | Default |
|---|---|---|---|
| reviewBrief | Yes | Structured review brief synthesized from conversation context | |
| consultantId | No | Review type. Currently all reviews use an independent panel of analysts. | |
| additionalContext | No | Additional context (e.g., {artifact: 'pull request', stakes: 'pre-merge gate'} or {domain: 'tax law', jurisdiction: 'Germany'}) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations declare readOnlyHint=false and openWorldHint=true; the description reinforces and enriches this by disclosing that the work is sent to 'a panel of reviewers running on another company's AI model' — material third-party data egress an agent cannot infer from the annotations. It does not cover cost, latency, or retention of submitted text, which are the remaining relevant traits for an external submission.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Front-loaded with the purpose in the first clause, then trypophobia-free scoping and exclusions. The mid-length 'Covers requests such as' list is justified because it broadens retrieval matching, though the description runs three dense paragraphs and could have trimmed one example cluster.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a single-required-param, nested-object tool with no output schema, the description fully carries the load: it explains the input's completeness requirement, the external reviewer panel, the shape of the returned verdict, and the boundary cases. Nothing an agent needs to decide or invoke correctly is missing.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the baseline is 3, but the description adds real semantics: it explains that 'the review covers the complete text submitted' and that an 'unabridged copy of the work gives a complete result', which directly steers how to populate responseUnderReview. It also implies the brief is synthesized from conversation context, matching the nested reviewBrief object.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb+resource ('independent review of AI-generated work') and enumerates the concrete artifact types it accepts (answer, code, PR, plan, analysis, draft). The second paragraph goes further and names the actual return payload (verdict, inaccuracies, gaps, follow-up questions, report card), so an agent knows exactly what it gets back.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Provides both an inclusion list ('is this right', hallucination check, fact-check, adversarial review, gate review, sign-off before merging) and an explicit 'Not for' exclusion block covering claim verification, document Q&A, AI-detection, agent evaluation, and post-hoc discussion of a returned review. It also clarifies re-use ('a revised draft can be reviewed again'), which removes a common ambiguity about whether re-submission is allowed.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
submit_review_request1 field changedInput schema / properties / reviewBrief / properties / responseUnderReview / descriptionPrevious value: -"The AI-generated work under review (your answer, or the code, PR summary, plan, analysis or draft), as-is. Do not omit, summarize, or condense any content."New value: +"The complete, unabridged text of the AI-generated work under review (your answer, or the code, PR summary, plan, analysis or draft)."submit_review_request4 fields changedInput schema / properties / additionalContext / descriptionPrevious value: -"Additional context (e.g., {petType: 'Golden Retriever', petAge: '2 years'})"New value: +"Additional context (e.g., {artifact: 'pull request', stakes: 'pre-merge gate'} or {domain: 'tax law', jurisdiction: 'Germany'})"Input schema / properties / reviewBrief / properties / objective / descriptionPrevious value: -"What the user is trying to achieve"New value: +"What the user is trying to achieve (e.g., 'merge PR #42 safely', 'submit this manuscript', 'decide whether to refinance')"Input schema / properties / reviewBrief / properties / responseUnderReview / descriptionPrevious value: -"The specific AI response the user wants reviewed. Provide the response text as-is — do not omit, summarize, or condense any content."New value: +"The AI-generated work under review (your answer, or the code, PR summary, plan, analysis or draft), as-is. Do not omit, summarize, or condense any content."Input schema / properties / reviewBrief / properties / userConcerns / descriptionPrevious value: -"User's specific concerns about the response"New value: +"What the user is worried about (e.g., 'the retry logic', 'whether the statistics section holds up')"submit_review_requestIndependent fact-checking for AI answers: a verdict and sources for every claim.
Multiple AIs peer-review and debate your question, then return one fact-checked answer.
Free fact-checks, papers, source vetting, plus verified AI pricing, comparisons, guides, and tools.
Checks a claim at its primary source and grades it. A person reviews every answer.
Add one secure layer between your agents and this server.
