Broke to Built

Convert SRT to VTT or VTT to SRT, strip timings to a plain text transcript, and resync cues by any offset. Paste text or pass up to 50 file URLs. Handles optional VTT hours, comma vs dot decimals and NOTE blocks. $0.005 per file; files with no cues are free.

Live TLS handshake for up to 200 hosts per run: issuer, SANs, days-to-expiry, full chain walk, negotiated protocol, self-signed/expired diagnosis with exact error. $0.0008 per host, no start fee, unreachable hosts never charged — cheaper than measured incumbents ($0.001-$0.035).

Search Stack Overflow or any Stack Exchange site and get one row per question: score, answer count, views, tags, author reputation and ISO dates. Tag filters, bulk searches, no API key needed. $0.005 per question; empty searches are free.

Scrape Steam game reviews: recommended or not, full text, helpful and funny votes, and the author's playtime AT THE TIME OF REVIEW in hours. Filter by language, sentiment and sort. Bulk apps, no API key. $0.002 per review.

Scrape any Substack newsletter by handle OR custom domain: post title, author, publish date, subtitle, cover image and word count, one row per post. Optional full text. Bulk publications. $0.004 per post; empty feeds are free.

Scrape any public Telegram channel: post text, publish date, view counts, photo/video flags, forward and reply markers, and a direct link. One row per post, pages past the 20 a page shows, bulk channels. No API key or login. $0.01 per post.

Round-trip case conversion done right: camelCase, PascalCase, snake_case, kebab, CONSTANT_CASE, Title, Sentence, dot.case, slug — with correct acronym splitting (parseHTMLParserOutput -> parse_html_parser_output). 200 texts/run. $0.0004 per text all nine cases; junk never charged.

Regex extraction of 11 structured entity types from up to 100 texts per run: emails, phones, URLs, IPs, dates, times, money, percentages, hashtags, mentions + Luhn-checked ALWAYS-MASKED cards. Not model NER — exact, private, in-process. $0.0008 per text all types, junk entries never charged.

Detect the language of up to 200 texts per run: ISO 639-3 code, English name, confidence and ranked alternatives (franc trigram model, in-process, texts never leave the run). $0.0005 per text — half the measured incumbent ($0.001); too-short snippets never charged. For routing and dataset splitting.

Score up to 100 texts per run with Flesch, Flesch-Kincaid, Gunning Fog, SMOG, ARI and Coleman-Liau + consensus grade, audience label and concrete fix-it tips. $0.0004 per text all six formulas, no start fee, word-free inputs never charged — vs $0.0005-$0.015 measured incumbents.

Sentiment analysis for any text via API. Input: text string(s). Output: JSON with sentiment label (positive/negative/neutral), normalized score, comparative score, and the driver words. Lexicon-based, no API key. Cheap pay-per-result: $0.01 per text analyzed.

Exact counts for any text: characters with and without spaces, words, unique words, sentences, paragraphs, syllables, longest words, per-word and per-sentence averages, and lexical diversity. Unicode-correct. Up to 200 texts per run. $0.02 per text analyzed.

Any timestamp shape in (unix s/ms auto-detected, ISO, "now") — every form out: unix, ISO 8601, RFC 2822, DST-correct IANA timezone rendering, real ISO week numbers, day-of-year, leap flag, relative time. $0.0004 per conversion (cost-plus floor, no paid rival measured); bad inputs never charged.

Scrape any public Tumblr blog: post title, id, publish date, full tag list, word count and a reblog flag, one row per post. Accepts a blog name, a tumblr.com URL or a custom domain. Bulk blogs, no API key. $0.002 per post.

Physical-unit conversion by API: kg/kilos/kilogramme all resolve, temperature uses real formulas (212F = exactly 100C, release-tested), every result includes the value in ALL sibling units of its category, float noise stripped. $0.40 per 1,000 conversions.

Parse up to 1,000 User-Agent strings per run: browser + version, OS, device, engine, CPU, and isBot verdict covering search crawlers, AI crawlers (GPTBot, ClaudeBot...), HTTP libraries and monitors. $0.0004 per UA, no start fee, junk strings never charged. For log analysis and traffic audits.

Generate up to 10,000 ids per batch: UUID v4, time-sortable UUID v7, ULID, NanoID, 8-char short codes — CSPRNG-backed, several kinds per run. $0.0004 per batch whatever its size vs $0.01/start measured incumbent. Unknown types never charged, never silently substituted. Release-tested formats.

Check a page's AMP setup: is it AMP, the linked amphtml URL, required AMP markup, and validation issues. Bulk. $0.0005 per page. Failed fetches are recorded free.

Scan a page for broken images: every img src/srcset HEAD-checked for 404s/errors/non-image types, plus missing-alt count. Bulk. $0.002 per page, cheaper than paid broken-link checkers; failed page fetches free.

Verify a page's character-encoding: HTTP vs meta charset, http-equiv, XML, BOM, and mismatch/mojibake risks. Bulk. $0.0005 per page. Failed fetches are recorded free.

Extract HTML comments from any page and flag sensitive ones - TODO/FIXME, passwords, API keys, tokens, internal notes. Single or bulk. $0.001 per page. Failed fetches are recorded free.

Find every icon a site declares - link icons, apple-touch, mask-icon, web-manifest icons, favicon.ico fallback - with sizes/types and a best-icon pick. Single or bulk. $0.0004 per page, cheaper than paid favicon actors; failed fetches free.

Extract a page's H1-H6 heading structure as an ordered outline with SEO issue flags (missing/multiple H1, skipped levels). Single or bulk. $0.0015 per page, a fraction of paid SEO-audit actors; failed fetches are free.

Estimate page weight and get performance hints from one fetch (no headless): HTML/transfer size, script/style/image/font counts, render-blocking, fixes. $0.001 per page, far below Lighthouse actors; failed fetches free.