Skip to content

fix(poll): preserve feedback in delivery output - #260

Closed
Julian-Dasilva wants to merge 1 commit into
kunchenguid:mainfrom
Julian-Dasilva:fm/lavish-upstream-228
Closed

fix(poll): preserve feedback in delivery output#260
Julian-Dasilva wants to merge 1 commit into
kunchenguid:mainfrom
Julian-Dasilva:fm/lavish-upstream-228

Conversation

@Julian-Dasilva

@Julian-Dasilva Julian-Dasilva commented Aug 20, 2026

Copy link
Copy Markdown
Contributor

Summary

The ack-before-consume protocol is deliberately deferred to #168, per maintainer guidance.

Fixes kunchenguid#228; deliberately does not touch kunchenguid#168 ack-before-consume.
@greptile-apps

greptile-apps Bot commented Aug 20, 2026

Copy link
Copy Markdown

Confidence Score: 5/5

The PR appears safe to merge with no actionable correctness, security, or repository-rule issues identified.

The actual CLI serialization is covered by an end-to-end test that confirms feedback precedes the bulky snapshot, while the documentation changes accurately describe the existing consume-on-delivery behavior and remain aligned with the repository vision.

Reviews (1): Last reviewed commit: "fix(poll): prioritize feedback in delive..." | Re-trigger Greptile

@kunchenguid

kunchenguid commented Aug 20, 2026

Copy link
Copy Markdown
Owner

Speaking as Kun's firstmate:

Holding this. The change matches the #228 guidance: emit prompts (then artifact_failures) before the bulky dom_snapshot, tell agents that delivery consumes the response, and leave ack-before-consume to #168. I reviewed the diff (no security issue) and approved first-time fork CI.

Required CI re-check: PR must be raised via no-mistakes is red — the body does not contain the ## Pipeline / Updates from git push no-mistakes signature the required check looks for (the HTML attestation is present, but that is not what CI greps). build-and-test is green. Please re-raise so that signature lands on this PR; I will not add the marker.

This is not a duplicate of #261 (that one is #229 presence/send). They do both edit the same README Agent-presence paragraph, so they should not land as a pair without rebasing the one that remains.

@Julian-Dasilva

Copy link
Copy Markdown
Contributor Author

Content unchanged; re-raised through the required no-mistakes pipeline in #264.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants