Filter up to 25 records against up to 10 yes/no criteria you supply

Filter up to 25 records against up to 10 yes/no criteria you supply. Each record comes back with keep / review / drop, a reason naming the criterion it failed, a 0-10 score and the per-criterion probabilities. Never drops a record it could not read or could not judge: those come back as review or keep, flagged. Built on a decision model rather than a chat model, so it is a flat per-call price with no token accounting on your side.

Paid test badge: not yet. The checks above are free: we call the tool without paying and read the payment request it sends back. The Verified badge needs paid calls whose answers match the promised output, and nobody can buy a badge.

Endpoint

POST https://x402-production-0f93.up.railway.app/v1/filter

Our checks, last 30 days

Example input (from the provider)

{
  "body": {
    "criteria": [
      {
        "key": "in_europe",
        "question": "The company operates in Europe."
      },
      {
        "key": "makes_things",
        "question": "This is a company that manufactures or automates physical production, not a service business."
      }
    ],
    "mode": "all",
    "records": [
      {
        "company": "Nordfjord Packaging AS",
        "contact": "S. Halvorsen",
        "country": "Norway",
        "id": "1"
      },
      {
        "company": "Bright Ideas Consultancy",
        "contact": "",
        "country": "UK",
        "id": "2"
      }
    ]
  },
  "bodyType": "json",
  "method": "POST",
  "type": "http"
}

Promised output schema (from the provider)

{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "properties": {
    "input": {
      "additionalProperties": false,
      "properties": {
        "body": {
          "properties": {
            "criteria": {
              "items": {
                "properties": {
                  "key": {
                    "description": "Short snake_case name. Appears in the result as failed_<key>.",
                    "pattern": "^[a-z][a-z0-9_]*$",
                    "type": "string"
                  },
                  "question": {
                    "description": "One yes/no statement about a single record, phrased so that TRUE means you want the record. Narrow and factual works; 'is this a good fit' does not.",
                    "type": "string"
                  },
                  "weight": {
                    "description": "Only used when mode is \"score\". Defaults to 1.",
                    "type": "number"
                  }
                },
                "required": [
                  "key",
                  "question"
                ],
                "type": "object"
              },
              "maxItems": 10,
              "minItems": 1,
              "type": "array"
            },
            "dropBelow": {
              "description": "score mode only. Below this a record is dropped; between this and threshold it is marked review. Defaults to half the threshold.",
              "type": "number"
            },
            "mode": {
              "default": "all",
              "description": "all: every criterion must hold, each as its own gate so a drop is attributable. any: at least one must hold. score: the weighted mean must clear the threshold.",
              "enum": [
                "all",
                "any",
                "score"
              ],
              "type": "string"
            },
            "records": {
              "description": "Any shape. At most 4000 characters per record.",
              "items": {
                "type": "object"
              },
              "maxItems": 25,
              "minItems": 1,
              "type": "array"
            },
            "threshold": {
              "description": "In all/any mode, how certain one criterion must be to count as true (0-1, default 0.5). In score mode, the 0-10 score a record must reach (default 5).",
              "type": "number"
            }
          },
          "required": [
            "records",
            "criteria"
          ]
        },
        "bodyType": {
          "enum": [
            "json",
            "form-data",
            "text"
          ],
          "type": "string"
        },
        "method": {
          "enum": [
            "POST"
          ],
          "type": "string"
        },
        "type": {
          "const": "http",
          "type": "string"
        }
      },
      "required": [
        "type",
        "method",
        "bodyType",
        "body"
      ],
      "type": "object"
    },
    "output": {
      "properties": {
        "example": {
          "properties": {
            "meta": {
              "properties": {
                "bundle": {
                  "type": "string"
                },
                "bundle_version": {
                  "type": "string"
                },
                "elapsed_ms": {
                  "type": "integer"
                },
                "engine": {
                  "type": "string"
                },
                "failures": {
                  "type": "integer"
                },
                "items": {
                  "type": "integer"
                },
                "model_cost_usd": {
                  "description": "What the call cost the server in model tokens. Published so you can see the markup you are paying.",
                  "type": "number"
                },
                "questions_sha256": {
                  "description": "Hash of the exact question wording used, so a result is reproducible.",
                  "type": "string"
                }
              },
              "type": "object"
            },
            "results": {
              "items": {
                "properties": {
                  "confidence": {
                    "enum": [
                      "high",
                      "medium",
                      "low"
                    ],
                    "type": "string"
                  },
                  "error": {
                    "description": "Present only when this one item could not be judged. Its verdict is then the safest rung, never a discard.",
                    "type": "string"
                  },
                  "id": {
                    "description": "The record's id field, or its index.",
                    "type": "string"
                  },
                  "rationale": {
                    "description": "The per-signal probabilities behind it.",
                    "type": "string"
                  },
                  "reason": {
                    "description": "A single label explaining the verdict.",
                    "type": "string"
                  },
                  "rules_applied": {
                    "description": "Which deterministic safety gates fired.",
                    "items": {
                      "type": "string"
                    },
                    "type": "array"
                  },
                  "score": {
                    "description": "0-10.",
                    "type": "number"
                  },
                  "verdict": {
                    "description": "One rung of the ladder for this operation.",
                    "type": "string"
                  }
                },
                "type": "object"
              },
              "type": "array"
            }
          },
          "type": "object"
        },
        "type": {
          "type": "string"
        }
      },
      "required": [
  

This page as JSON