Certify the ORIGIN of a piece of content via the provenance-receipts service. Returns an Ed25519-signed receipt committing to a SHA-256 hash of the content, your caller-attested generator_metadata, and a service-set timestamp.

PROVES: the content is unmodified (hash) and the receipt was issued by the holder of the service's signing key. DOES NOT PROVE: it does not independently verify generator_metadata — the receipt proves you CLAIMED that metadata at certification time, not that a specific model actually produced the content. This is proof-of-origin/tamper-evidence, not AI-detection and not a truth guarantee.

Note: /v1/certify is the paid action; when the underlying service has x402 payments enabled it may require a micropayment. This wrapper forwards the request as-is.

Verify a provenance receipt against its content via provenance-receipts. Re-hashes the content and checks the Ed25519 signature; returns { valid, details }. valid is true only if BOTH the content hash matches the receipt AND the signature is valid under the service's public key. Free. A 200 response means verification ran — always read valid (false means the content was modified or the receipt was altered/forged).

Score content against the quality-gate service's PUBLISHED rubric and get an Ed25519-signed score receipt. The rubric scores four dimensions — clarity, completeness, internal consistency, and obvious-error freedom (0–25 each, summing to 0–100) — plus flags from a closed vocabulary. Read the exact rubric with get_quality_rubric.

WHAT THIS IS: a reproducible score AGAINST OUR PUBLISHED RUBRIC (vX). WHAT IT IS NOT: a measure of absolute truth, an external/third-party standard, or a fact-check — 'obvious-error freedom' catches errors evident from the text or common knowledge, not external verification.

Optional target_score (0–100) enables 'no pass, no pay': if the score is below your target, no receipt is issued and (when payments are on) no charge is made — the response returns the failing breakdown with receipt: null.

Note: /v1/score is the paid action (one Claude API scoring pass per call) and may require an x402 micropayment when the underlying service has payments enabled.

Verify a quality score receipt against its content via quality-gate. Re-hashes the content and checks the Ed25519 signature, which covers the score itself — so a forged or altered score fails. Returns { valid, details }. Free. A 200 response means verification ran — always read valid.

Fetch the published Quality Gate rubric (markdown) that score_quality grades against, along with its version. Read this to understand exactly what a quality score means and does not mean. Free; takes no input.