canonical: https://jentic.com/apis/found17.com/found17 # Found17 API The Found17 API runs deterministic, source-backed checks on public web pages: it extracts a normalized product record from a product page, evaluates whether supplied sources establish a claim, and runs reproducible HTTP, HTML, SEO, and accessibility checks on a page. It compares product records across pages, returns a source-hashed snapshot with caller-controlled change checks, and manages time-bounded public-source watches. It also lists available services and quotes their cost before a call runs. ## For AI agents Extract normalized product records from public pages, verify whether sources establish a claim, and run reproducible SEO, HTML, and accessibility checks on a webpage. Compare product records, snapshot pages with change detection, and manage public-source watches. ## Scope Does not host content, crawl private pages, or store a persistent dataset. Use for source-backed extraction, verification, and audits of public web pages only. ## Capabilities - Extract a normalized product record from a public product page without inventing missing fields - Evaluate whether supplied public sources establish a claim and return the evidence - Run reproducible HTTP, HTML, SEO, and accessibility checks on a public webpage - Turn deterministic page checks into a ranked, evidence-backed website action list - Compare source-backed product records from multiple pages in one call - Return a bounded, source-hashed snapshot of a page with caller-controlled change checks - Create, check, and remove time-bounded public-source watches ## Use cases ### Agent Web Verification An AI agent connected through Jentic can extract a product record from a page, verify a claim against supplied sources, and audit the page for SEO and accessibility issues, without a developer wiring the Found17 API by hand. Found17 returns evidence-backed results without fabricating citations, so the agent acts on grounded output. Example prompt: Extract a product record from a product page and verify a specific claim about it against supplied sources ### Product Data Extraction Catalog and comparison tools pull structured product data from arbitrary product pages. Found17 extracts a normalized product record from one page and compares records from several pages in a single bounded call, returning source-backed fields. Example prompt: Extract product records from three product pages and compare them in one call ### Website Audit Teams check a page for reproducible HTTP, HTML, SEO, and accessibility issues and want a prioritized list of fixes. Found17 runs the checks and turns them into a ranked, evidence-backed website action list. Example prompt: Run web QA checks on a webpage and return a ranked action list of issues to fix ## Key endpoints | Method | Path | Description | | --- | --- | --- | | POST | `/v1/web-to-product` | Extract a normalized product record from a page | | POST | `/v1/verify-claim` | Evaluate whether sources establish a claim | | POST | `/v1/web-qa` | Run HTTP, HTML, SEO, and accessibility checks | | POST | `/v1/website-audit` | Produce a ranked website action list | | POST | `/v1/compare-products` | Compare product records across pages | | POST | `/v1/source-snapshot` | Return a source-hashed page snapshot | | GET | `/v1/quote` | Quote a service's cost | ## Key resources - **Web extraction** — Extract a normalized product record from a page and compare records across pages - **Verification** — Evaluate whether supplied public sources establish a claim - **Web QA and audit** — Run reproducible page checks and produce a ranked action list - **Snapshots** — Return a source-hashed snapshot of a page with caller-controlled change checks - **Source watches** — Create, get, delete, and check time-bounded public-source watches - **Services** — List available services and quote their cost before a call runs ## AI readiness This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use. - **Score:** 50 / 100 - **Maturity:** Foundational - **Dimensions:** - Foundational Compliance: 82 / 100 - Developer Experience & Jentic Compatibility: 64 / 100 - AI-Readiness & Agent Experience: 27 / 100 - Agent Usability: 94 / 100 - Security: 50 / 100 - AI Discoverability: 56 / 100 - **View full report:** https://jentic.com/apis/found17.com/found17/scorecard - **How the score is calculated:** https://docs.jentic.com/reference/api-readiness-framework/overview/ - **More about the dimensions:** https://docs.jentic.com/reference/api-readiness-framework/specification/#dimensional-model-overview ### Score it yourself Every API in the directory is allowlisted, so you can re-score it with no key required. - **Score your own API:** https://jentic.com/scorecard.md - **Scoring CLI agent skill:** https://github.com/jentic/jentic-api-scorecard/blob/main/skills/jentic-api-scorecard/SKILL.md ```sh npx @jentic/api-scorecard-cli score ``` ## Why Jentic - **Setup:** Wiring the Found17 API by hand means handling its pay-per-call model, authorizing each call with an x402 payment signature or a Stripe billing account header, and shaping each page request yourself. Through Jentic you install once, import Found17 from the API Directory, store the credential once, and your agent calls it. - **Permission scoping:** Found17 carries the target page URL in the request body, so a rule bounds which operations your agent may call rather than which page: you can allow the extraction and audit operations and leave source-watch deletion out. You choose the operations it may call, so destructive ones like deleting a source watch are not included unless you add them. - **Credential handling:** Your Found17 credential is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context. - **Discovery method:** Agents search Jentic by intent such as 'extract a product record' or 'audit a website', and Jentic returns the matching Found17 operation with its input schema so the agent calls the right endpoint without browsing the reference docs. ## Related APIs - **Firecrawl** — Web crawling and extraction API that turns pages into structured data. - **ApiFlash** — Screenshot API for capturing a visual record of a page. - **Urlbox** — Rendering API for screenshots and PDFs of web pages. ## FAQ ### What authentication does the Found17 API use? Found17 uses a pay-per-call model declared as API key headers in its OpenAPI spec: agents authorize each call with an x402 USDC payment signature on Base mainnet sent in the `PAYMENT-SIGNATURE` header, while human or business usage pairs a Stripe billing account id in the `X-Found17-Account-Id` header with a billing-mode header. Through Jentic the credential is stored encrypted by your own Jentic One instance and injected at call time. ### Can I extract a product record with the Found17 API? Yes. The web-to-product operation extracts a normalized product record from a public product page without inventing missing fields, and the compare-products operation evaluates records from several pages in one bounded call. ### Can I limit what my agent is allowed to do with the Found17 API? Yes. Found17 carries the target page URL in the request body, so write a rule that allows only the extraction and verification operations, such as extracting a product record and running web QA, letting the agent read and check pages but not delete a source watch. Every call it makes is logged. ### Is there a Found17 MCP server? You do not need an MCP server to give your agent the Found17 API. Jentic connects it directly from the API Directory: import Found17, store your credential once, and the agent calls its operations by intent without loading another server's tool definitions into its context. ### What are the rate limits for the Found17 API? The OpenAPI spec does not specify numeric rate limits for the Found17 API; usage is metered per call through its pay-per-call model. Check the Found17 documentation at https://found17.com for current limits and pricing. ### How do I audit a web page with the Found17 API through Jentic? Search Jentic for an intent like 'audit a website', and it returns Found17's web QA and website-audit operations with their input schemas so your agent can run the checks and get a ranked action list. To run it on your own infrastructure, install Jentic One from its GitHub repo.