UGC Reactions from Real Creators
Real creator library videos at $11 and images at $8. Custom production from $25/video with a $250 minimum. Commercial licensing included.
翻訳待ち
リポジトリ情報はまだ収集していません。
No updates detected yet. GitHub releases will appear here.
The maker's introduction consists of the line above. Read the repository README for more details. Repository README ↗
Loading the full README.
Read the full README ↗Webflow app by Silit Labs. "Every AI. One deck." Keyword by keyword, every week, it checks whether seven AI engines (ChatGPT, Google AI Overviews, Google AI Mode, Gemini, Perplexity, Claude, Grok) name and cite a brand. Each keyword is asked as two customer questions on the chat engines and as the search itself on Google AI Overviews and AI Mode (12 answers a check). For each keyword it shows yes, sometimes or no, the answers, the rivals named instead and a fix plan (best page, title and description, FAQ schema, CMS drafts, a new page draft when no page fits) applied to Webflow in one click. Related keywords come from the AI's own searches and DataForSEO Labs (keywords where Google's AI Overviews already cite the site, related searches, monthly volumes).
Live address (planned): https://aiseodeck.silitlabs.com
api/index.js: one serverless entry point (pages, OAuth, fixes, cron). vercel.json rewrites every app path to it.api/_lib/engines.js: ChatGPT (OpenAI Responses API + web_search), Google AI Overviews and AI Mode (DataForSEO SERP API), Gemini (Google Search grounding), Perplexity (Sonar, switching by itself to the Agent API that replaces it), Claude (Messages API + web search tool), Grok (xAI Responses API + web_search). An engine is on when its key is set. Free plan: ChatGPT, AI Overviews, Perplexity; the others show as locked columns.api/_lib/analyze.js: brand and rival detection in answers (case, accents, dashes and Turkish letters ignored), citations, share of voice, score (50% named, 30% cited, 20% place).api/_lib/insights.js: "Where rivals win" gaps and "How AI describes you" (sentiment, words, claims worth checking; one small LLM call per run).api/_lib/keywords.js: keyword cleaning, matching a keyword to the best page, DataForSEO Labs (volumes, related searches, keywords cited in AI Overviews).api/_lib/jobs.js: runs are queues of keyword question x engine tasks in Redis; a run covers all keywords (weekly) or some (a keyword checked on demand, keywords just added). The cron tick and the open site page both work through them; each engine has its own parallel limit; failed calls are retried up to 3 times.api/_lib/audit.js + fetcher.js: reads live pages, robots.txt, llms.txt, sitemap.xml (only the site's own hosts, private addresses refused).api/_lib/fixes.js + llm.js: suggestions (questions, meta, FAQ answers), schema building and merging, llms.txt.api/_lib/billing.js + polar.js: Polar subscriptions. Only Polar products named "AI SEO Deck ..." (or with metadata app=aiseodeck) are used. Free trial: every checkout of an account that never had a trial or a subscription gets trial_interval: day, trial_interval_count: TRIAL_DAYS (default 7, 0 turns it off); the others get allow_trial: false. Back from checkout, /app/billing/return reads the checkout and its subscription so the plan is on before the webhook arrives./auth?plan=solo&interval=monthly -> Webflow -> /app/start (today $0, first charge date) -> Polar checkout -> /app/billing/return -> site setup.test/app.js: end-to-end test with mock Webflow, Redis, AI engines, Resend and Polar. Run npm test.| Plan | Price | Sites | Keywords | Checks |
|---|---|---|---|---|
| Free | $0 | 1 | 3 | monthly, 3 engines (ChatGPT, AI Overviews, Perplexity), no Apply to Webflow |
| Solo | $19/mo, $190/yr | 1 | 10 | weekly, all 7 engines, Apply, 10 keyword checks on demand a month |
| Studio | $49/mo, $490/yr | 5 | 30 | as Solo with 30 checks on demand, client report links with agency name |
| Agency | $149/mo, $1490/yr | 20 | 100 | as Studio with 100 checks on demand |
New keywords get a first check right away, up to the plan's number of keywords in new keywords a month.
Every paid plan starts with a 7-day free trial (card taken by Polar, charged on day 8 unless cancelled; one trial per account, and Polar's "Prevent trial abuse" also blocks a second trial with the same email or card). A Solo trial costs at most about $4 in AI calls.
Rough AI cost per answer (Nov 2026 list prices): ChatGPT about 1.1 to 2 cents (web search $10 per 1,000 calls), AI Overviews 0.4 and AI Mode 0.4 (DataForSEO live), Gemini about 0.2 within the 5,000 free grounded prompts a month and about 3 after ($14 per 1,000 search queries), Perplexity about 0.5 to 0.8, Claude about 1.2 to 3 (web search $10 per 1,000), Grok about 2.5 to 4 (web search $5 per 1,000 calls plus tokens). One keyword check (12 answers) is about 19 cents (12 to 26), so about $0.82 a keyword a month with weekly checks. Worst case per month (every keyword, every check on demand and every new keyword check used): Solo about $12, Studio about $36, Agency about $120. DataForSEO Labs lookups cost 1.2 cents plus 0.012 cents per keyword. MAX_DAILY_CALLS (default 1500, at most about $22 a day) stops all checks for the day once reached; work continues the next day.
This is the beginning of the README. Read the rest at the original. Read the full README ↗