Data x402 offerings

Compare Separate provider confidence bands from unknown confidence values. against a bounded caller-supplied previous snapshot and return exact change witnesses. Baseline is caller supplied and is ne…

Recall remembered text by similarity to a query (ranked by cosine similarity), not by exact key. Returns the top-k matches with scores. The retrieval half of the wallet-scoped semantic memory.

Composite transformation: extracts structured data, entities, tables, and stats from text. One call replaces 4 endpoints.

Composite transformation: extracts structured data, entities, tables, and stats from text. One call replaces 4 endpoints.

Composite transformation: extracts structured data, entities, tables, and stats from text. One call replaces 4 endpoints.

Deterministic entity extraction from text (no LLM): emails, URLs, EVM addresses, tx hashes, ENS/basenames, dollar/USDC amounts, ISO dates, percentages. Returns structured JSON with counts.

Use this when an agent has an older and a newer LLM catalog snapshot and needs a deterministic migration diff before changing routing or budgets. Rows are matched by vendor plus model, id or name. The result separates added, removed and changed models and reports field-level differences for input, output and cached-input price, context window and maximum output tokens.

Compares two submitted sample sets feature by feature and returns deterministic drift metrics. Continuous features are binned on the combined value range and receive PSI, Jensen-Shannon divergence and a Kolmogorov-Smirnov statistic; categorical features receive PSI and Jensen-Shannon divergence over exact category labels. The response names every feature, its sample counts, the metric values, the threshold used, and the reason a feature was flagged.

Compares two submitted sample sets feature by feature and returns deterministic drift metrics. Continuous features are binned on the combined value range and receive PSI, Jensen-Shannon divergence an…

Compare supplied structures with explicit salt, protonation and stereochemical handling. For dataset reconciliation; no empirical identity or purity confirmation.

Get a machine-ready demographic, economic and housing profile for a U.S. geography, with Census uncertainty and peer benchmarks.

Extract named entities from text using Claude Haiku — people, organizations, locations, dates, emails, URLs, money amounts, and products — returned as structured typed arrays so agents can pull structured signals out of unstructured text in one call.

Split text into overlapping chunks for RAG ingestion — by characters or words, with configurable chunk size and overlap. Deterministic (same input → same chunks), returns each chunk with its index and start/end offsets. No API key, no LLM. Pairs with /v1/embeddings for a full chunk→embed pipeline.

Transform a source object into a target schema — matches fields by alias and structure deterministically, uses an LLM only for semantic mapping when needed, and supports mapping hints, transforms, and defaults, so agents can reshape data between systems with per-field confidence and a clear report of what mapped and what didn't.

Turn unstructured text into structured JSON matching a caller-supplied schema using Claude Haiku — the agent declares the fields and types it wants, and gets back populated JSON with values pulled from the text and coerced to the right types, so agents can structure any prose into the exact shape their pipeline expects.

Parse a PDF to text and Markdown: send a PDF url or base64 file, get clean Markdown with paragraphs rebuilt from the text layer, per-page character offsets for citing pages, metadata (title, author, dates, page count) and the page numbers that are scanned images with no text layer. Up to 50 pages per call. Text layer only, no OCR: an image-only PDF returns an uncharged 422.

Conform messy or inconsistent JSON to a target schema using Claude Haiku — renames keys, coerces types (string→number, "true"→boolean, etc.), and fills missing fields with null, returning JSON that matches the exact shape the caller specified so agents can normalize data from varied sources into one consistent structure.

Event extraction / event parsing — turn any caption, announcement, listing, or page text into a normalized calendar event via Claude Haiku. Resolves relative dates ("this Saturday 7pm") against post time + timezone, handles all-day/multi-day, and returns title, start/end, venue, address, city, price, organizer, url. Purpose-built for structured event data (no schema to define); pairs with /event-classify and /calendar/ics.

Extract tabular data from messy text or HTML using Claude Haiku — detects columns and rows in unstructured content and returns clean structured JSON (columns + rows) so agents can turn pasted tables, HTML tables, and delimited text into usable data in one call.

Normalize any messy money string into a typed decimal amount plus ISO 4217 currency — handles symbols, locale separators ($1,234.56 vs €1.234,56), magnitude suffixes (1.2M), accounting notation, and natural-language amounts — so agents can clean money values before converting or storing them, with no external data and a deterministic fast path.

Validate data against a supplied schema — checks required fields, nested objects, arrays, enums, string formats, numeric ranges, lengths, and additional-property rules deterministically, with optional type coercion and repair, returning precise errors with JSON paths so agents can verify structured data before acting on it.

Extract structured data from any unstructured text into your own JSON Schema — structured-data / information extraction, text-to-JSON, LLM data enrichment. You supply the schema; the LLM returns a matching object. Works for contacts, invoices, events, product specs, medical records, legal clauses, resumes — any shape. Returns the extracted object plus token usage; unfound fields are omitted or null.

Transform a source object into a target schema — matches fields by alias and structure deterministically, uses an LLM only for semantic mapping when needed, and supports mapping hints, transforms, an…

Event extraction / event parsing — turn any caption, announcement, listing, or page text into a normalized calendar event via Claude Haiku. Resolves relative dates ("this Saturday 7pm") against post …