docs: package PromiseProof for judge-first evaluation#3
Merged
Conversation
Add JUDGE_START_HERE.md, a 60-second judge path that covers the recorded walkthrough, the in-browser challenge with its expected states, the authority split, and the local test commands. Lead the README first screen with the line that GPT-5.6 investigates, Codex repairs, and neither decides PASS, and link the judge guide. Point the Action examples in the README, ADOPTION.md, and the action README at the submission-rc-03 release tag.
Replace em dashes with plain punctuation in the user-visible text on the /verify/ page and the recorded walkthrough, and in a couple of source comments. Text only, with no behavior change. The hosted verifier suite still passes 11 of 11.
Add a Challenge the proof call to action in the header and hero linking /verify/?judge=1, and add a one-line thesis under the headline: GPT-5.6 investigates, Codex repairs, neither decides PASS. No redesign and no new sections.
Describe GPT-5.6 as ranking hypotheses and selecting one allowlisted replay rather than proposing the diagnosis, and state that verification runs in the browser after the page loads, with no backend API, evidence upload, or model call in the verdict path.
AlexPaiva
added a commit
that referenced
this pull request
Jul 21, 2026
Judge-first documentation, landing challenge call to action, and copy polish.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What this adds
Judge-first documentation. No product or verifier change.
JUDGE_START_HERE.md: a 60-second judge path covering the recorded walkthrough, the in-browser challenge with its expected states, the authority split (who may investigate, propose source changes, approve the exact patch, and decide PASS), and the local test commands.ADOPTION.md, and the action README pin thesubmission-rc-03release tag.Not changed
No evaluator, schema, adapter, binding, report, verifier, or Action code. No landing, walkthrough, or hosted-site change, so no redeploy is needed. Built from
submission-rc-03(6d25ad9).