Genesis402

Extract JSON fields from text

genesis402_extract_json

Pull user-defined fields from unstructured text as JSON, returning null for anything not stated and never inventing values. Turns emails, invoices, contracts, or listings into structured data.

Instructions

Paid ($0.006 USDC). Pull the fields you define out of unstructured text as JSON. Anything the text does not state comes back null, never invented. Use to turn emails, invoices, contracts or listings into structured data. To get a web page's text first use genesis402_web_extract. The field descriptions steer the extraction, so include units and formats (e.g. "ISO date", "number in USD"). Returns one JSON object with exactly your keys, each holding the extracted value or null. Payment: call once without payment_signature to get the exact price quote (nothing is charged or signed), then call again with the signed payment to receive the result and settlement details.

Input Schema

TableJSON Schema

Name	Required	Description
`text`	Yes	The source text to read, up to 16,000 characters. Trim longer input first.
`fields`	Yes	Up to 30 fields: an object mapping each output key to a short description of what it should hold, including unit or format, e.g. { "invoice_total": "total amount due in USD, number", "due_date": "ISO date" }.
`payment_signature`	No	Optional. An x402 v2 payment you signed for this call's quote (the PAYMENT-SIGNATURE header value). Omit to get the price quote first; nothing is charged without it.

Output Schema

TableJSON Schema

Name	Required	Description
`mode`	Yes	QUOTE_ONLY: price quote only, nothing signed or charged. PAID: the call was paid and result holds the data. FREE: the endpoint answered without payment.
`paid`	Yes	true only when a payment was settled for this call.
`price`	No	The Base USDC quote. Present on QUOTE_ONLY and on PAID calls made by a local payer; null when the rail offered no Base USDC lane or the caller signed the payment.
`result`	No	PAID or FREE: the endpoint's JSON result.
`resource`	No	QUOTE_ONLY: the full URL of the paid resource.
`how_to_pay`	No	QUOTE_ONLY: how to sign and resend the call.
`settlement`	No	PAID: decoded PAYMENT-RESPONSE settlement (transaction hash, network, payer), or null if the rail sent none.
`what_you_get`	No	QUOTE_ONLY: the rail's description of what the payment buys.
`payment_required_b64`	No	QUOTE_ONLY: the raw base64 PAYMENT-REQUIRED header, to sign with any x402 v2 client.
`also_accepted_networks`	No	QUOTE_ONLY: every network the 402 challenge accepts.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedv0.3.4Sep 29, 2026

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "$schema": "http://json-schema.org/draft-07/schema#",
+  "additionalProperties": false,
+  "properties": {
+    "also_accepted_networks": {
+      "description": "QUOTE_ONLY: every network the 402 challenge accepts.",
+      "items": {
+        "type": "string"
+      },
+      "type": "array"
+    },
+    "how_to_pay": {
+      "description": "QUOTE_ONLY: how to sign and resend the call.",
+      "type": "string"
+    },
+    "mode": {
+      "description": "QUOTE_ONLY: price quote only, nothing signed or charged. PAID: the call was paid and result holds the data. FREE: the endpoint answered without payment.",
+      "enum": [
+        "QUOTE_ONLY",
+        "PAID",
+        "FREE"
+      ],
+      "type": "string"
+    },
+    "paid": {
+      "description": "true only when a payment was settled for this call.",
+      "type": "boolean"
+    },
+    "payment_required_b64": {
+      "description": "QUOTE_ONLY: the raw base64 PAYMENT-REQUIRED header, to sign with any x402 v2 client.",
+      "type": [
+        "string",
+        "null"
+      ]
+    },
+    "price": {
+      "anyOf": [
+        {
+          "additionalProperties": false,
+          "properties": {
+            "atomic": {
+              "description": "Price in USDC atomic units (6 decimals), as a decimal string.",
+              "type": "string"
+            },
+            "network": {
+              "description": "CAIP-2 network id of the lane, eip155:8453 (Base).",
+              "type": "string"
+            },
+            "payTo": {
+              "description": "Address the payment settles to.",
+              "type": "string"
+            },
+            "usd": {
+              "description": "Price in US dollars.",
+              "type": "number"
+            }
+          },
+          "required": [
+            "atomic",
+            "usd",
+            "payTo",
+            "network"
+          ],
+          "type": "object"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "description": "The Base USDC quote. Present on QUOTE_ONLY and on PAID calls made by a local payer; null when the rail offered no Base USDC lane or the caller signed the payment."
+    },
+    "resource": {
+      "description": "QUOTE_ONLY: the full URL of the paid resource.",
+      "type": "string"
+    },
+    "result": {
+      "description": "PAID or FREE: the endpoint's JSON result."
+    },
+    "settlement": {
+      "description": "PAID: decoded PAYMENT-RESPONSE settlement (transaction hash, network, payer), or null if the rail sent none."
+    },
+    "what_you_get": {
+      "description": "QUOTE_ONLY: the rail's description of what the payment buys.",
+      "type": "string"
+    }
+  },
+  "required": [
+    "mode",
+    "paid"
+  ],
+  "type": "object"
+}

Changed2 schema fields changedv0.3.3Sep 29, 2026

changedInput schema / properties / fields / description

Previous value: -"The fields to extract: an object mapping each field name to a short description, e.g. { \"invoice_total\": \"total amount due in USD\", \"due_date\": \"ISO date\" }."New value: +"Up to 30 fields: an object mapping each output key to a short description of what it should hold, including unit or format, e.g. { \"invoice_total\": \"total amount due in USD, number\", \"due_date\": \"ISO date\" }."

changedInput schema / properties / text / description

Previous value: -"The source text to read, up to 16,000 characters."New value: +"The source text to read, up to 16,000 characters. Trim longer input first."

Changed3 schema fields changedv0.3.2Sep 26, 2026

addedInput schema / properties / fields / description

Added value: +"The fields to extract: an object mapping each field name to a short description, e.g. { \"invoice_total\": \"total amount due in USD\", \"due_date\": \"ISO date\" }."

changedInput schema / properties / payment_signature / description

Previous value: -"Optional. An x402 v2 payment you signed for this call's quote (the PAYMENT-SIGNATURE header value). Omit to get the price quote first."New value: +"Optional. An x402 v2 payment you signed for this call's quote (the PAYMENT-SIGNATURE header value). Omit to get the price quote first; nothing is charged without it."

addedInput schema / properties / text / description

Added value: +"The source text to read, up to 16,000 characters."

First observedv0.3.1Sep 26, 2026

TDQS

A4.8/5.0

Behavior5/5

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

Annotations only cover readOnlyHint=false and openWorldHint=true. The description adds material behavior beyond that: the $0.006 USDC charge, the two-call quote-then-sign flow where nothing is charged without a signature, and the anti-hallucination guarantee that unstated fields return null. These are non-obvious traits that meaningfully affect how an agent calls the tool.

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?

Front-loaded with price and purpose, then use cases, sibling routing, parameter guidance, return contract, and payment flow. Every sentence carries information, though the combined payment and return explanation makes it denser than strictly necessary.

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?

Despite high complexity (nested object param, payment protocol, output schema), the description covers purpose, prerequisites, parameter crafting, return shape, and the full payment handshake. With an output schema present, it does not need to restate return values and correctly avoids that.

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 coverage is 100%, so the baseline is 3. The description goes further by explaining that field descriptions steer extraction and should carry units/formats, and it names the exact output contract ('one JSON object with exactly your keys'). That is real guidance beyond the schema's structural definitions.

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 ('pull the fields you define out of unstructured text as JSON') and distinguishes itself from genesis402_web_extract by naming that sibling as the upstream step. An agent can tell this is a schema-driven extraction tool, not a generic text fetcher.

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?

Gives concrete use cases (emails, invoices, contracts, listings) and explicitly routes to genesis402_web_extract when the source is a web page. It also prescribes the exact payment call sequence, so the agent knows the correct invocation pattern rather than guessing.

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