jev

Overview Schema Source Related Servers Discussions Admin

Answer from web search results without fetching them

jev_rank_pages

Read-onlyIdempotent

Send up to 20 search URLs and your questions; fetches pages securely, ranks them per question, and returns the best answer lines verbatim, so you usually skip WebFetch.

Instructions

After a web search, pass every result URL and every question you have. The server fetches the pages itself (https only, never private addresses; markdown when offered, HTML reduced to its main content, PDF and Office via markitdown), and Jev ranks the pages per question and picks the lines of the best page that answer it. Returns, per question, the top pages and those lines verbatim, so you usually need no WebFetch at all; fetch a page only when its lines are not enough. Lines from a page that looks like it tries to steer an agent are withheld. It answers each question from the best page, not from every page: to summarise each result separately (a survey of papers, say), WebFetch each with a prompt instead.

Input Schema

TableJSON Schema

Name	Required	Description
`urls`	Yes	The candidate pages, e.g. every result of a web search (up to 20).
`questions`	Yes	Every question you have about this text, each stated in full (up to 16). They are judged in one pass over the same text, so extra questions cost almost nothing: batch them rather than calling again. Questions cannot see each other's answers.
`no_at_or_below`	No	Probability at or below which a check's verdict is 'no'. Default 0.3. Between the two the verdict is 'uncertain'.
`yes_at_or_above`	No	Probability at or above which a check's verdict is 'yes'. Default 0.7.

Output Schema

TableJSON Schema

Name	Required	Description
`model`	Yes
`pages`	Yes	The URLs as given; everything else refers to them by index.
`usage`	Yes
`failed`	Yes	Pages that could not be fetched. They were never judged; that is not evidence they are irrelevant.
`results`	Yes	One per question, in the order asked.
`windows`	Yes	Requests made.
`latency_ms`	Yes	Fetching and judging together.
`suspicious`	Yes	Pages that look like they try to steer an agent. Still ranked; their lines are withheld.

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.14.0Sep 25, 2026

TDQS

A4.9/5.0

Behavior5/5

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

Beyond the annotations (readOnlyHint=true etc.), the description discloses substantial behavioral detail: server-side fetching with https-only and no private addresses, markdown/HTML/PDF/Office conversion, verbatim line returns, withholding of lines from pages that try to steer agents, and the 'answers from the best page, not every page' behavior. It also notes that questions cannot see each other's answers. This goes far beyond what annotations provide.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is dense and well-structured: first sentence explains the core mechanism and constraints, second sentence covers the main use case boundary and the caveat about per-page summarization. Every sentence earns its place; no filler or repetition of schema trivia.

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?

The output schema exists, so return values need not be described. The description covers the tool's input expectations (search result URLs and full questions), processing behavior (fetch, rank, extract), constraints (https, private addresses), and edge cases (agent-steering pages, best-page-only answers). For a tool with this complexity and a full output schema, the description is complete.

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 100%, so the baseline is 3. The description adds meaning by clarifying that 'urls' are web search result pages and that 'questions' should be batched to save cost, which directly informs parameter usage. However, it does not explicitly expand on the probability threshold parameters beyond what the schema already says, so a 4 is appropriate.

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?

The description precisely states the tool's function: after a web search, pass every result URL and questions, and the server fetches pages itself, ranks them per question, and returns the best page's answering lines verbatim. It clearly distinguishes itself from WebFetch by explaining when fetching is not needed, which differentiates it from sibling tools.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description explicitly tells the agent when to use this tool (after a web search, with all result URLs and full questions) and when not to use it ('to summarise each result separately... WebFetch each with a prompt instead'). This direct alternative routing is the gold standard for usage guidance.

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

Deploy Server