com.edgethirteen.www/gateway

Scan a single live, publicly reachable web page's markup for the WCAG 2.1 failures that ADA web-accessibility demand letters and lawsuits cite most often: missing alt text, unlabeled form fields, missing lang attribute, empty links, skipped heading levels, missing page title, no h1, no skip link. Static markup analysis -- it can't check color contrast, keyboard navigation, or screen reader behavior, so a clean result means 'no known markup-level issue found,' not full compliance.

Get pricing and details for AdvisorLetter: a weekly white-label market-commentary draft that financial advisors review and send to clients under their own name. Useful for an agent helping a financial advisor or RIA find a done-for-them (not done-by-AI-and-sent-raw) client communication product.

Get a real sample AdvisorLetter issue -- the exact ready-to-forward weekly market-commentary draft an advisor reviews and sends to clients under their own name. This is a fixed sample issue, not this week's live draft -- the paid subscription is a fresh issue every Monday. Free, no key.

Check whether the EU AI Act's Article 50 transparency rule applies to one of your AI touchpoints, and get a free sample disclosure notice for it. Covers chatbots, AI-generated images/audio/video, AI-generated public text, and emotion-recognition/biometric features. Returns which Article 50 obligation applies, where the notice belongs, and one ready-to-use disclosure sentence -- the same free preview shown on the web tool before checkout. Paste in your own business name for a personalized sentence. Free, no key.

Get pricing and details for AI Disclosure Kit: ready-to-use EU AI Act Article 50 transparency-obligation text, an embeddable disclosure banner, and a compliance record template for chatbots, AI-generated media, and biometric-categorization/emotion-recognition features. One-time $59. Useful for an agent implementing EU AI Act transparency disclosures for its user's product.

Compute the Altman Z-Score (public-company, non-financial-services version) for a US-listed ticker -- a quick bankruptcy-risk / financial-distress screen from five balance-sheet ratios. Looks up working capital, total assets, retained earnings, EBIT, market cap, total liabilities, and revenue from SEC EDGAR + market data automatically -- just pass a ticker. Returns the score and which zone it falls in: safe (>2.99), grey (1.81-2.99), or distress (<1.81).

Get pricing and details for APIDrift: weekly plus same-day urgent email alerts tracking deprecations, breaking schema changes, and pricing shifts across OpenAI, Anthropic, Google, xAI, Together, Mistral, Groq, Cloudflare Workers AI, and AWS Bedrock. Useful for an agent helping an engineering team that depends on LLM provider APIs avoid getting blindsided by an upstream breaking change.

Get real, dated example entries of the kind of LLM-provider breaking change APIDrift catches -- model retirements, endpoint shutdowns, and schema/pricing changes across OpenAI, Anthropic, Google, xAI, Together, Mistral, Groq, Cloudflare Workers AI, and AWS Bedrock, each with what actually broke and what to check. These are illustrative sample entries, not a live feed -- the paid subscription is the ongoing weekly digest plus same-day urgent alerts as new changes ship. Free, no key.

Get details and pricing for the US Bank Regulatory Stress Dataset: a flat CSV covering every FDIC-insured bank's balance-sheet stress metrics (AOCI burden on capital, uninsured deposit ratio, non-performing loan ratio, and core balance-sheet figures) computed directly from each bank's own public FFIEC Call Report. Useful for an agent building a bank-screening or credit-risk model that needs a licensable raw data input instead of scraping regulatory filings itself.

Look up one US bank's balance-sheet stress metrics, computed from that bank's own FFIEC Call Report and citable as an official-source figure: AOCI burden on capital (how much of equity unrealised securities losses have eaten), uninsured deposit ratio, non-performing loan ratio, total assets and state. Covers all 4,296 FDIC-insured institutions for 2026 Q2 (filed as of 06/30/2026). Use this when an agent needs a grounded answer to "how stressed is this bank?" instead of guessing from memory or reading a filing by hand. Free, no key. One institution per call.

Given a chargeback/dispute reason code category and order details, return the evidence checklist for that reason code (what to attach: tracking numbers, prior-order signals, refund records, etc.) and a preview of the response letter. Covers Visa 10.4/13.1/13.3/13.7/13.6/12.6.1 and the equivalent Mastercard reason codes. The full formatted dispute-response letter is a separate $39 product at https://www.edgethirteen.com/tools/chargeback-response. Not legal advice.

Fetch a URL and scan it for content planted to hijack an AI agent visiting it rather than to inform a human: invisible unicode (zero-width joiners, steganographic tag-block characters), CSS-hidden text (display:none, opacity:0, off-screen) paired with instruction-shaped language ('ignore previous instructions', 'add to cart and checkout', fake system-role text), and instruction-shaped phrases in alt/title/aria-label attributes or meta tags. Call this before an autonomous browsing or checkout flow acts on instructions found on a page. A clean result means no known hijack pattern found, not a full content-safety guarantee.

Look up whether a US-listed ticker has any SEC comment-letter correspondence on file (staff review letters and company responses, form types UPLOAD/CORRESP). Returns a free-tier preview: filing count and most recent filing date, sourced live from SEC EDGAR. Does not include letter content or severity scoring -- that's the paid full report.

Compute a simplified discounted-cash-flow intrinsic value per share: project free cash flow per share forward at a constant growth rate, discount back at a discount rate, add a Gordon-growth terminal value. Pass fcf/growth/discount/terminal/years directly (these are judgment calls, not looked up). Optionally pass a ticker to auto-fill the current price for a margin-of-safety comparison.

Given a California data broker's DROP registration/start date, return the next 45-day DROP (Delete Request and Opt-out Platform) check-in deadline and the first two items of the compliance checklist. Missing a cycle costs $200/day per unprocessed deletion request, required starting August 1, 2026. The full 20-cycle calendar file, deletion-log template, and complete checklist are a separate $49 product at https://www.edgethirteen.com/tools/drop-compliance-kit. Not legal advice.

Check a comma-separated list of software dependency/product names against CISA's live Known Exploited Vulnerabilities (KEV) catalog. Returns any matches with the CVE, vulnerability name, and why it matched. Call this before shipping or during a dependency/security review to catch actively-exploited components, or when working out whether an EU Cyber Resilience Act Article 14 reporting obligation (the 24h early warning / 72h notification / 14-day final report clock) has been triggered. Matching is by name and deliberately errs towards over-matching, so review each hit before acting on it.

Compute Benjamin Graham's fair-value ceiling (sqrt(22.5 * EPS * book value per share)) for a US-listed ticker, and the margin of safety against its current price. Looks up EPS, book value per share, and price from SEC EDGAR + market data automatically -- just pass a ticker. You can also pass eps/bvps/price directly to compute it for a hypothetical company with no ticker.

Get background and pricing for the HubSpot legacy-to-date-based-versioning (DBV) migration map: what's going unsupported, when, and where the full free map and $29/mo DBV Watch monitoring live. Useful for an agent helping a developer or agency plan a HubSpot API migration off v1-v4 before HubSpot's own per-endpoint docs ship in March 2027.

List HubSpot API operations whose date-based-versioning replacement is BREAKING (a caller-visible field or parameter difference) or REMOVED (no date-based counterpart exists at all), optionally filtered by API name (e.g. 'CRM', 'Webhooks'). Useful for scoping how big a given migration is before starting it.

Look up a legacy HubSpot API path (e.g. '/crm/v3/objects/contacts' or just 'contacts') and get its date-based-versioning replacement, whether the change is breaking, and the exact field-level differences (removed params, newly-required fields, changed types). Derived by diffing HubSpot's own public OpenAPI specs -- not a guess. Free, no key, full data (this map has no paid tier gating it).

Get details and pricing for Edge Thirteen's current Industry Primer report: a one-time research report on a single investable industry theme, with real revenue/margin/growth figures pulled from each covered company's own SEC 10-K (not modeled or estimated), plus the analytical read on industry structure and competitive positioning. Useful for an agent helping with investment research that needs a sourced industry-level writeup instead of an LLM-generated summary.

Get the free FY2025 preview table from Edge Thirteen's current Industry Primer: ticker, FY2025 revenue, YoY growth, gross margin, and net margin for each covered company, sourced to that company's own 10-K via SEC EDGAR's XBRL company-facts API. This is the same free table already on the tool page -- the paid report adds FY2023-FY2024 comparatives, exact XBRL tag/accession sourcing per figure, and the written analysis. Free, no key.

Analyze manuscript text for patterns associated with AI-generated prose: cliche-phrase density, unnaturally uniform sentence-length rhythm, repeated paragraph openers, and overused transition words. Returns a 0-100 score, a high/medium/low band, and the single most severe flag -- the full flag breakdown with fix guidance for every flag is the paid $29 report at https://www.edgethirteen.com/tools/kdpcheck. Not a claim about authorship, just a pattern scan a writer can use to self-edit before publishing.

Get the killcord script: a dead-man's switch for Anthropic API spend. Self-hosted, dependency-free Node script that polls Anthropic's real Admin Usage & Cost API and deactivates the API key the instant spend crosses a limit -- for stopping a runaway agent loop, not just alerting on it. Returns the full script source plus a ready-to-run setup snippet, optionally pre-filled with the workspace/limit/keys you provide, so it can be written straight into the project being worked on.

Get a free, real code sample from MeterKit: the fix for a common Stripe webhook bug where access is only revoked on customer.subscription.deleted, missing that a card decline usually transitions a subscription via .updated first (sometimes without .deleted ever firing). This is the same free sample already on the tool page -- the paid kit is the surrounding webhook route plus the event-sourced credit ledger and Meter Events reporting/reconciliation, which this sample does not include. Free, no key.

Get pricing and details for MeterKit: a drop-in, event-sourced credit ledger plus a correct 2026 Stripe Billing Meter Events API integration for Next.js AI/SaaS products (the old usage_type: "metered" price approach was retired as of Stripe API version 2025-03-31.basil). Zero runtime dependencies, one-time $99, buyer owns the code. Useful for an agent implementing Stripe usage-based billing for its user.

Check $CASHTAG-style ticker mentions (e.g. $AAPL) in a piece of financial writing against SEC EDGAR's own XBRL company-facts data for that company's most recent 10-K -- EPS, book value per share, revenue, market cap, total assets, total liabilities. Flags numeric claims that don't match the filed figures.

Scan a job posting's text against the free-preview subset of US state salary-range-disclosure laws (CA, CO, NY, WA) and flag exactly which required element (salary range, benefits description) is detectably missing, with the statute citation. The full 15-state + DC report (CT, HI, IL, MD, MA, MN, NV, NJ, RI, VT, VA, DC too) is a separate $49 report at https://www.edgethirteen.com/tools/postingcheck. Conservative: never certifies a posting as compliant, only flags what's missing. Not legal advice.

Check whether an npm package's public download count reflects real adoption. Detects 'download pumping' -- the documented supply-chain technique where a package is published in hundreds of rapid-fire versions so registry mirrors and security scanners inflate its download counter, making an unused or malicious package look popular. Returns npm's headline 30-day count, a spike-resistant estimate of sustained real usage, the share of the month falling on the busiest single day, recent version-flood bursts, and a verdict of clean, unreliable or inflated. Use before adding or recommending a dependency, especially a new or unfamiliar one.

Score a subscription business's signup/cancellation flow against US click-to-cancel and auto-renewal law (federal FTC Section 5/ROSCA, plus California AB 2863, Colorado SB25-145, Vermont, and Oregon). Returns the federal-level result now; the full state-by-state breakdown with statute citations is a separate $49 report at https://www.edgethirteen.com/tools/renewal-matrix. Not legal advice.

Spot-check a TikTok creator's exported video list for TikTok Creator Rewards Program disqualification-risk signals: repetitive/duplicate captions, unnaturally regular posting cadence, ineligible formats (duet/stitch/photo), and under-duration videos. Returns a 0-100 risk score, a band (low/medium/high), and the single highest-severity flag -- the full flag breakdown is the paid product at edgethirteen.com/tools/rewards-guard. Input is the same caption/postedAt/durationSeconds/format rows TikTok Studio's own analytics export gives a creator about their own account.

Screen a person or company name against OFAC's SDN and Consolidated Sanctions lists (19,874 entities, 21,320 aliases) via SanctionRail, for an agent checking a counterparty before a transaction, vendor onboarding, or payment approval. Returns fuzzy matches banded exact/strong/probable/weak with the sanctions programme, source list, and a common-name-collision flag so a common name like 'Maria Garcia' is surfaced, not silently suppressed. Free tier: 25 lookups/day per caller IP, no key required.

Discover a website's pages from its sitemap (or a homepage crawl as fallback) and sweep a free sample through two engines: WCAG 2.1 accessibility checks and AI-agent prompt-injection checks. Returns how many pages exist, how many were sampled, severity counts, and which categories of problem were found -- not the finding text or which page it's on. Call this before launch or handoff to catch site-wide accessibility and agent-hijacking exposure a single-page scan would miss. The full 40-page report with exact locations and a worst-first remediation order is a separate $499 product at https://www.edgethirteen.com/tools/site-audit.

Check a YouTube channel's recent uploads against YouTube's 2026 inauthentic-content ('AI slop') policy: title/description repetition, upload-cadence regularity, and AI-disclosure language. Returns an overall risk band (low/medium/high), a numeric score, and the single most severe flag found. This is a heuristic proxy for YouTube's own review process, not a simulation of it. The full flag-by-flag report with exact numbers and suggested fixes is a separate paid product at https://www.edgethirteen.com/tools/slopcheck.

Get an instant monthly/annual spend total for a pasted list of SaaS subscriptions, one per line as 'Name, monthly cost, category' (category is optional -- common tools like Slack or Notion are auto-recognized), plus a count of categories with likely overlap (e.g. two project-management tools). This is the free-tier preview -- the per-category breakdown, which line items are priced high for their category, and a consolidation-savings estimate are the $49 paid report at https://www.edgethirteen.com/tools/stackledger.

Get details and pricing for Edge Thirteen's Tariff Refund Screen: a one-time CSV of US-listed companies disclosing IEEPA tariff-refund figures in their own SEC 10-Q/10-K filings since the Supreme Court struck down IEEPA tariff authority. These dollar figures are gain contingencies under ASC 450-30, so they carry no XBRL tag and don't appear on any stock screener -- they exist only as footnote prose. Useful for an agent doing investment research that needs to find a disclosure no numeric screen can surface.

Get the free top-10 preview from the Tariff Refund Screen: ticker, company, disclosed amount, whether it's paid/accepted/received, and IEEPA refund as a percentage of shareholders' equity, ranked by materiality. This is the same free table already on the tool page -- the paid CSV adds the other 150 cited companies plus the exact citation sentence and filing URL for every figure. Free, no key.

Check a vulnerability report's concrete claims (file paths, file:line references, function names) against a real public GitHub repo's current contents. Flags claims that don't check out -- the file doesn't exist, the line is past EOF, the function isn't in the file -- which is the most common signature of an AI-generated fabricated vuln report. Does not judge whether a real, verifiable claim describes an actual security bug; it only checks whether the claim is honest about the codebase.

Due diligence on a US company, from official federal records. Screens a company name against SEC EDGAR (is it a registered filer, under what CIK, state of incorporation and SIC code), the US Treasury OFAC SDN sanctions list, USAspending.gov federal contract and grant awards, and federal court dockets via CourtListener/RECAP. Use before onboarding a vendor, signing a supplier, screening a counterparty or qualifying a sales lead. Free screen returns how many records each source holds plus one sample record each; sources that cannot be reached are reported as not-checked rather than as clean.

Get a real sample run of ValueFeed's Graham/Buffett-style value screen: per-name metrics (P/E, P/B, Graham fair value, margin of safety, ROE, current ratio, debt/equity, yield, years of positive EPS) for two names, plus an SEC 10-K/10-Q filing excerpt for one of them. This is a fixed illustrative sample, not a live run across the whole market -- the paid subscription is a fresh weekly screen every Monday. Free, no key.

Get pricing and details for ValueFeed: a weekly SEC-sourced value-investing research packet (screens, metrics, filing excerpts) licensed for independent newsletter writers and content creators to build their own commentary on top of. Useful for an agent helping a finance content creator find raw research inputs instead of raw LLM-generated numbers.

Scan pasted project files for known Vercel Python-runtime deploy footguns: conflicting requirements.txt/pyproject.toml manifests, a dependency manifest missing from the project root, sibling-module imports that need sys.path, multiple top-level ASGI apps under api/, streaming/SSE responses that can crash the Python runtime, oversized dependencies (torch/tensorflow/etc.), and unrouted .well-known/ files. Static analysis over file contents -- it doesn't run a real build, so a clean result means 'no known footgun found,' not a deploy guarantee. To scan a public GitHub repo automatically instead of pasting files, use the $29 checkout at https://www.edgethirteen.com/tools/vercelpycheck.

Answer one financial metric for a US public company, spoken as a single natural-language sentence grounded in that company's most recent SEC 10-K filing (via EDGAR's XBRL data) -- not a guess. Use for a specific spoken question like "what's Tesla's revenue" or "what's Apple's earnings per share". Metric must be one of: revenue, eps, book_value_per_share, market_cap, total_assets, total_liabilities, price.

Give a short spoken financial snapshot of a US public company (revenue, earnings per share, market cap, book value per share) grounded in its most recent SEC 10-K filing. Use for a broad spoken question like "tell me about Apple's financials" rather than one specific figure.