Changelog.

Product history for the Santos Website Intelligence API. Latest version: 2.10.0. Machine-readable version and contract data: /version.

2.10.0 — 2026-07-26

Santos Feed Parser (GET/POST /v1/feed, $0.003 USDC): one feed URL to normalized JSON — detects RSS 2.0, Atom, and JSON Feed through the SSRF-guarded fetcher; returns feed metadata plus up to 50 items; non-feed targets return 422 and never settle.
Santos Link Map (GET/POST /v1/links, $0.003 USDC): one HTML page to a categorized link map — every link (max 200) with an internal/external kind plus topic tags (docs, pricing, api, careers, social, feed) and per-category counts.
Santos Summarizer (POST/GET /v1/summarize, $0.033 USDC): one HTML page to a Claude-generated structured summary (title, summary, key_facts, entities, word_count) with an optional focus steering prompt; non-HTML targets return 422 and never settle.
New MCP tools feed_parse, link_map, and summarize: each validates the target and returns the canonical x402 HTTP handoff for its paid endpoint — MCP never executes or settles the paid call.
Buyer examples: buy-feed.js, buy-links.js, buy-summarize.js. Discovery surfaces updated: OpenAPI, llms.txt, capabilities manifest, /api service manifest, and /docs.

Verified-email free audits: GET /api/audit/free runs the full Quick Intelligence audit once per day per verified email (6-digit code flow at /api/leads/verify/*, token valid 30 days). The browser widget now captures email up front.
Public reports + leaderboard: opted-in audit scores are published at /reports/<domain> with a top-50 leaderboard at /reports.
Agent-Ready badge: GET /v1/badge?url=… returns a free SVG shield of a site's latest public score — embed it in any README.
llms.txt generator: GET /v1/llms-txt/demo drafts a standards-shaped llms.txt from a one-page sample (verified-email free tier, shared 1/day quota); tool page at /llms-txt-generator.
Signed reports + POST /v1/verify: audit reports now carry an HMAC-SHA256 signature; verify any report free at /v1/verify (30/hour) or the /verify page.
CI recipe: examples/agent-readiness-ci.yml and .sh gate pull requests on the Agent Readiness score (~$0.075 USDC/run via x402); guide at /ci.

Batch Quick Intelligence Audit is now flat $0.50 USDC for up to 50 URLs (was $0.10 / 10 URLs) — $0.01 per URL at full capacity.

Batch Quick Intelligence Audit (POST /api/audit/batch, $0.10 USDC): up to 10 URLs in one payment, per-URL failure isolation; settles only when at least one audit succeeds.

Free Agent Readiness demo (GET /api/agent-readiness/demo) — shared 1/day/IP quota, same result shape as the paid audit; added to OpenAPI, llms.txt, the service manifest, and the MCP audit_agent_readiness handoff.
Demo-exhaustion 429 responses now carry for_humans (card checkout pointer) and retry_after across all five demo endpoints; the browser widget offers email capture (POST /api/leads) when the daily quota is spent.
First-party analytics now persist to a durable store; payment_completed is recorded for both x402 and Stripe settlements.
New trust surfaces: /status, /changelog, /version. robots.txt simplified to open-by-default for agent discovery.
Fix: human card checkout now charges the advertised $5 (was $19 server-side).

Structured Extraction (POST /v1/extract/structured, $0.08 USDC): LLM-extracted JSON validated against the caller's own JSON Schema; settles only when output validates.
New extract_structured_data MCP tool with a shared-quota free preview.

Screenshot & PDF Render (GET /v1/screenshot, $0.01 USDC): real-Chromium PNG/JPEG/PDF, desktop or mobile, optional full-page capture.

Safe Fetch (GET/POST /v1/fetch, $0.002 USDC): raw text body of one public URL through the SSRF-guarded fetcher, with redirects, headers, and timing.

Page-to-Markdown Extraction (POST /v1/extract, $0.005 USDC): readability-isolated main content as clean Markdown with title, links, and metadata.

x402 Bazaar discovery extension with per-route input/output schemas; agent-friendly robots.txt Allow rules for paid endpoints.