Skip to main content
Glama

NS Pros MCP Gateway

noirlab-datalab__prediction_calibration

REALIZED win rate by shown price and minutes elapsed for short-horizon crypto up/down markets, net of taker fee, read from our own settled-outcomes archive, not from live prices. Answers the one question no venue will: "a side shown at 82 with five minutes left, how often did it ACTUALLY win?" Returns a grid of (price bucket x minute elapsed) -> n, realized_win_rate_pp, mean_shown_price_pp, calibration_gap_pp (realized minus shown, zero means the book is calibrated), breakeven_win_rate_pp (the s

Example tool call

This call uses example arguments supplied in the tool’s input schema.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "asset": "btc",
      "venue": "polymarket",
      "minute": 5,
      "horizon": "15m"
    },
    "name": "noirlab-datalab__prediction_calibration"
  }
}

Input Schema

NameRequiredDescriptionDefault
assetNobtc (default) | eth | sol | xrp | bnb. Coverage differs by venue: polymarket and polymarket-us carry btc/eth/sol/xrp, Kalshi carries btc/eth/bnb.
venueNopolymarket | polymarket-us | kalshi | all (default polymarket). "all" returns one grid per venue, never a merged one.
minuteNoMinutes elapsed since window open, 0-based. Omit for the full grid; pass 5 for the "five minutes in" row only, which is also the cheapest query.
horizonNo15m (default). 5m | 1h | 1d are accepted but only 15m is collected today, so another value returns coverage.windows_total 0 rather than an error.
bucket_ppNoPrice-bucket width in percentage points: 5 | 10 (default) | 20 | 25 | 50. Must divide 100, so the last bucket is never narrower than the rest.
side_conventionNo"favorite" (default) folds each window onto whichever side was the favorite, the frame published claims about early favorites use. "yes" keeps the venue own YES/Up side.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4/5.0
Behavior4/5

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

With no annotations, the description carries the full load and does substantial work: it discloses data provenance (own settled-outcomes archive, not live markets), the fee treatment (net of taker fee), and the meaning of key output semantics such as calibration_gap_pp being zero when the book is calibrated. It does not describe cost, latency, or pagination behavior, and the trailing clause about breakeven_win_rate_pp is cut off mid-sentence.

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

Conciseness3/5

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

The lead sentence is well front-loaded with resource and scope, but the body is one long run-on sentence that piles field definitions together, and the text terminates abruptly inside the definition of breakeven_win_rate_pp ('(the s'), leaving a visibly broken clause. Useful content, but the structure and truncation cost it.

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?

Six optional parameters with no output schema mean the description must explain return values, and it mostly does — enumerating the grid dimensions and three of the four derived metrics. However, the truncation leaves breakeven_win_rate_pp undefined, which is a real gap for an agent that must interpret the grid before reporting results.

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 documents every parameter in detail, including coverage differences by venue, the 0-based minute semantics, bucket widths, and the side_convention frame. The description adds almost nothing about parameters — its added value is in output-field semantics rather than inputs — so the baseline of 3 applies.

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 pair with precise scope: realized win rate by shown price and minutes elapsed for short-horizon crypto up/down markets, net of taker fee, drawn from a settled-outcomes archive. It explicitly names what it is not ('not from live prices') and frames the concrete question it answers, so an agent can distinguish it from any live-price tool at a glance.

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?

The use case is made concrete through the example question ('a side shown at 82 with five minutes left, how often did it ACTUALLY win?') and the explicit exclusion of live-price framing, which tells the agent when this tool rather than a pricing tool applies. It stops short of naming an alternative tool or stating when-not-to-use, so it falls below a 5.

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.