caleb-brown/temper
1
0
Fork 0
Code Issues 36 Pull requests Projects Releases Packages Wiki Activity Actions

Publish advisory status and a managed summary #16

New issue
Open
opened 2026-07-14 22:32:21 -04:00 by caleb-brown · 0 comments
caleb-brown commented 2026-07-14 22:32:21 -04:00
Owner
Copy link

What to build

Developers receive one current, advisory Forgejo review summary and an execution status that accurately describes completion and coverage without acting as a merge verdict.

Acceptance criteria

  • Temper publishes pending status when review execution begins and a terminal execution state when publication completes or fails.
  • Findings do not turn the execution status into approval, request-changes, or an authoritative merge verdict.
  • Exactly one managed summary exists per pull request and is updated for the latest published revision.
  • The summary identifies the reviewed SHA, time, profiles, findings, usage when configured, exclusions, and coverage.
  • Complete, partial, failed, and no-findings outcomes use distinct language; no-findings explicitly disclaims approval and security guarantees.
  • Repeating publication for the same desired result is idempotent.
  • The full review-journey test proves the Forgejo-visible status and summary against a deterministic model result.

Blocked by

  • #15 — Produce a bounded review result
## What to build Developers receive one current, advisory Forgejo review summary and an execution status that accurately describes completion and coverage without acting as a merge verdict. ## Acceptance criteria - [ ] Temper publishes pending status when review execution begins and a terminal execution state when publication completes or fails. - [ ] Findings do not turn the execution status into approval, request-changes, or an authoritative merge verdict. - [ ] Exactly one managed summary exists per pull request and is updated for the latest published revision. - [ ] The summary identifies the reviewed SHA, time, profiles, findings, usage when configured, exclusions, and coverage. - [ ] Complete, partial, failed, and no-findings outcomes use distinct language; no-findings explicitly disclaims approval and security guarantees. - [ ] Repeating publication for the same desired result is idempotent. - [ ] The full review-journey test proves the Forgejo-visible status and summary against a deterministic model result. ## Blocked by - [#15](https://git.caleb-brown.dev/caleb-brown/temper/issues/15) — Produce a bounded review result
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
dev
t3code/hygiene-review
t3code/configurable-llm-timeout
fix/openai-non-streaming-completions
t3code/test-incorrect-changes
No results found.
Labels
Clear labels   
ready-for-agent
Fully specified and ready for an implementation agent

  
wayfinder:grilling
Human-in-the-loop product or domain decision

  
wayfinder:map
Canonical Wayfinder map

  
wayfinder:prototype
Human-in-the-loop concrete prototype

  
wayfinder:research
Agent-driven primary-source research

  
wayfinder:task
Task that unblocks a Wayfinder decision

No labels
ready-for-agent
wayfinder:grilling
wayfinder:map
wayfinder:prototype
wayfinder:research
wayfinder:task
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Blocks
#17 Publish safe inline findings
caleb-brown/temper
Depends on
#15 Produce a bounded review result
caleb-brown/temper
Reference
caleb-brown/temper#16
No description provided.