Repository navigation
Local Publish decides its own next step, measured first - #130
Merged
Merged
Conversation
- Measurement first: a labelled set from the stream's records (failed checks, reviews with findings, CI failures of PRs #91–#124) measured over three deciders — Jev via ai.decision.decide, critique-then-verdict gpt-6-luna, critique-then-verdict claude-opus-5-5 — on agreement, abstention at the chosen minConfidence, latency and cost; recorded in Plan 316's measurement section (yarn decisions:measure-loop) - After a failed check: one choice among recheck_on_timeout, fix_commission, operator_install_mismatch; onUndecided = today's failure, which wakes Helmsman; the fix brief assembled from the check's output tail and the failing guard's name, no prose author - After a review with findings: one yes_no per finding and a choice among fix_before_merge, publish_as_is, ask; minConfidence high, onUndecided = the fix round; the fix brief assembled from the findings that passed - The push gate, the scan and the posture untouched (316 D9); tests route planted check outputs and finding lists through each route and the undecided branch; Plan 315 §2 gains the decision, Plan 316's status names the first production consumer
…tems are entries, the comment matches the paths
…, the fix round waits for the approval and pushes nothing declined, the measure script survives its own history
…comments stop telling stories
…dle names its prerequisite, and the after-check decision waits for its measurement
…art another publication
… the last round's, and the pull request names each finding once
…was measured on, and a round hands on findings that fit
…ound reads the review's own findings
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.
Round 2 on the Local Publish decision slice, all seven items from the node's note: the per-finding each step removed from Publish Local Changes (engine capability and tests stay on ai.decision.decide; the fix brief carries every finding); the fix-or-not decision stays on Jev at 0.8 with undecided = today's path; the after-check decision stays at 0.8 with the timeout-recheck fallback; CI remedy recorded at 20% in Plan 316 §7 and 321 P1 kept on when rules; the measured tables recorded in Plan 316 §7; review 2d65f9d4's findings fixed (republish-fix's gate checks the approval decision, fix-findings depends on approve-push, decline-as-answer tests cover judged findings and a fix round); the five decisions-measure-loop.ts defects fixed (404 skip, inline: decoded locally, 429 backoff, --allow-escape-sequences, gh-404 skip). 21 files, applies clean against 2eaf74b, on the kept session.