No separate app. No sign-ups for participants. Tag @winwithpickr in a reply on X or use /giveaway in Telegram — the bot picks winners right there in the conversation. Seeded randomness, Solana-anchored proof, fully inline.
Two modes. Same verification.
Reply to any post with @winwithpickr or use /giveaway in Telegram.
Entries collected, filtered by your conditions, winner selected via seeded PRNG.
Result appears right in the conversation. No redirect, no external site. Verification link included for the curious.
Include a question for your audience to answer.
Participants reply with their best guesses before the deadline.
Reply with the correct answer to trigger scoring.
Claude evaluates each prediction. Closest answer wins.
Seeded PRNG with Solana-anchored proof. Anyone can independently reproduce and verify the result.
LLM-scored predictions with semantic matching. Sports auto-close at the final buzzer via live ESPN schedules, crypto and market questions close at their target time, and earlier picks on time-sensitive contests get a small bonus to reward conviction. Fairest tie-breaking for every contest type.
Runs inside the conversation. No external links for participants, no app to install, no sign-up. Tag, pick, done.
Replies, retweets, follows, hashtags, account age. Set requirements per giveaway.
Group DMs, webhooks, auto-deliver codes, claim forms, access grants. Configure via DM.
Shuffling, hashing, and verification logic is public. Audit it yourself.
Start free. Upgrade when you hit limits.
|
Free
$0
|
Popular
Pro
$19/mo
|
Business
$49/mo
|
|---|
On X, tag @winwithpickr on any tweet and describe what you want in plain English — the bot understands natural language. On Telegram, send /giveaway in your group. The bot collects entries, picks winners using seeded randomness, and replies with the result and a verification link.
Yes. A random seed is generated and its SHA-256 hash is anchored to Solana before the entry pool is even fetched. After the pick, the result hash is also anchored on-chain. The algorithm is open-source, so anyone can independently verify any result.
Predict mode lets you ask your audience a question. Participants reply with their predictions. When you give the correct answer, Claude AI scores every entry for accuracy (0-100) and the closest answer wins. Ties are broken with the same seeded random draw as random mode. Learn more →
No. Everything happens inline — the bot replies directly in your thread or group. No web app, no sign-up, no leaving the platform.
On X, if the natural language parser can't extract your intent, the bot automatically falls back to the structured command parser. You can also use structured commands directly if you prefer precise control.
Yes — 3 picks per month on the free tier. Pro ($19/mo) gives you 10 picks with larger pools and no watermark. Business ($49/mo) adds scheduled picks, fraud filters, and more. See pricing for details.