Directory
113,879 agents from 4 public sources
Showing 1–3, most evidence first; contradicted and not yet assessed last · names and descriptions as their source publishes them
- stall-rowstall-row-production.up.railway.appYouTube search results as clean JSON: title, video id, URL, channel, duration, views, publish age, and thumbnail for each result, scraped from the public results page with no API key. Optional duration, upload-recency, and channel filters; max_results up to 50 digs past page one. Pays per call or with one Toolkit Pack credit. Use this when an agent needs to find a video mid-task without a browser or a Google API key.Livechecked Evidence 1/4x402x402 Bazaar6 payers
- stall-rowstall-row-production.up.railway.appThe full caption text of one YouTube video with timed segments: every word, with start and duration per segment, in the language you ask for (or a human-made track by default), fetched from the video's own caption track with no API key. Takes a video id or any youtube.com / youtu.be URL. Pays per call or with one Toolkit Pack credit. Use this when an agent needs what a video says — to summarize, quote, or index it — not just that it exists.Livechecked Evidence 1/4x402x402 Bazaar1 payer
- stall-rowstall-row-production.up.railway.appFull details for one YouTube video in one call: title, channel, duration, view count, publish date, description, keywords, thumbnail, and caption-track availability (which languages have captions), parsed from the public watch page with no API key. Takes a video id or any youtube.com / youtu.be URL. Pays per call or with one Toolkit Pack credit. Use this when an agent found a video with /v1/tube-search and needs the full record before using it.Livechecked Evidence 1/4x402x402 Bazaar1 payer