Loading deck…
Ship Page / Deck
Omni Verify
Print view: one slide per page. Choose Save as PDF in the print dialog.
01 · Creation
Kane CLI Hack Day
Omni Verify
Kane CLI checks every promise in an AI-written post before it ships and after every release. Failed promises are repaired from what Kane saw; only passing posts publish.
Marketing promises and the real product drift apart silently. AI growth agents make it worse: they write promises at scale, and they learn from the numbers those promises produce. A wrong number teach
01 / 05 · Omni Verify
02 · See it work
Kane CLI Hack Day
See it work
- 1The OpenAI agent (gpt-5.4-mini) drafts a post from the app's real page and a Kane Explore map of the site, and turns every promise into a check: the link's campaign tag, every sentence against the page, every promised feature.
- 2Kane runs headless with --agent in real Chrome and stores what it sees: the hand-off href before clicking, true or false per phrase, a feature's output. My code judges those values.
- 3On FAIL, the agent repairs the promise from what Kane saw (for drift 1: the App Store link with the campaign's own ct), and Kane checks again.
02 / 05 · Omni Verify
03 · The harness
Kane CLI Hack Day
The intelligence behind shipping
The OpenAI agent (gpt-5.4-mini) drafts a post from the app's real page and a Kane Explore map, and turns every promise into a check. Kane CLI runs headless with --agent in real Chrome and stores what it sees; my code judges those values. On FAIL the agent repairs the promise from what Kane saw and Kane checks again. Publish stays locked until Kane's latest result passed that exact version. After a release, Re-verify replays Kane's recording with kane-cli testrun, and Audit past posts checks old posts against today's site.
- 01Agent drafts post + checks
- 02Kane checks in real Chrome (--agent)
- 03Code judges Kane's stored values
- 04Agent repairs, Kane re-checks
- 05Publish gate (PASS + human confirm)
04 · Proof
Kane CLI Hack Day
What Kane actually verified
- Team-reportedOn scribo.cc/?ct=threads-oct24 Kane read the App Store button before clicking: ct=site-en-hero, not the campaign's ct=threads-oct24. FAIL (KR-63). The agent rebuilt the link from what Kane saw: PASS (KR-68). Same FAIL then PASS in all 5 full loops.run 63 · errored · RUNS.md rows KR-63 and KR-68 (Kane run_end: passed; recorder shows ERRORED, ticket H-0007). Other loops: KR-13/14, KR-15/16, KR-27/28, KR-74/75.
- VerifiedAssurance loop: Kane designed 2 tests from the campaign copy and both passed on the repaired link (excludes ct=site-en-hero, reaches Scribo's App Store page). Coverage 75%; the other 25% (installs credited to the campaign) Kane marked unmeasurable from a browser.run 46 · passed · commit 414c939395 · Kane run KR-46 (kanehack report -- kane-cli testrun t-2 t-3), source-verified pass on commit 414c939.
- VerifiedRe-verify replayed Kane's own recording of the passing campaign and passed (KR-85; earlier KR-77). Only then did Publish to Threads unlock, and a real post went out through the gate: threads.com/@scribo.voca/post/De4-yPnATF5
05 · Outcome
Kane CLI Hack Day
What improved
Kane CLI checks every promise in an AI-written post before it ships and after every release. Failed promises are repaired from what Kane saw; only passing posts publish.
https://github.com/DainoJung/omni-verify · https://www.loom.com/share/428499f73e57490c818743d7e2de3d77 · Dain Jung