docs(kpis): count approvals as clean in K5 and drop the skip-review advice

ryangr0 commented

2026-10-03 11:50:30 +00:00

Owner

K5 treated every submitted review as a change, so an approval lowered the
clean-merge rate, and the page told the owner to merge without a formal
approval to keep the proxy honest. That rewarded skipping review. Ploeg
already stores each review's verdict, reviewer, head SHA and time
(pull_request_reviews, migration 0023) and each push with its pusher
(pull_request_events, migration 0033).

K5 now counts a pull request as changed only when a review requested
changes, or someone other than the pull request's author pushed to the
branch, between the awaiting_review settle and the merge. Approvals and
comment-only reviews never lower it. The skip-review sentence is gone, the
data gap lists only what is still missing (GitLab pushes name no pusher,
the 500-event activity bound, a missing author, ForgeBots being
configuration), and D1 is narrowed to the GitLab pusher.

The new K5 query was checked against Ploeg's embedded PostgreSQL with all migrations and 7 seeded pull requests (3 clean of 7 as expected). The Loop dashboard panel in webgrip/homelab-cluster must be updated separately (filed on the Homelab Roadmap).

Verified with mise run verify on the pinned toolchain (all gates passed).

Ticket: https://vikunja.webgrip.dev/tasks/1750

🤖 Generated with Claude Code

K5 treated every submitted review as a change, so an approval lowered the clean-merge rate, and the page told the owner to merge without a formal approval to keep the proxy honest. That rewarded skipping review. Ploeg already stores each review's verdict, reviewer, head SHA and time (pull_request_reviews, migration 0023) and each push with its pusher (pull_request_events, migration 0033). K5 now counts a pull request as changed only when a review requested changes, or someone other than the pull request's author pushed to the branch, between the awaiting_review settle and the merge. Approvals and comment-only reviews never lower it. The skip-review sentence is gone, the data gap lists only what is still missing (GitLab pushes name no pusher, the 500-event activity bound, a missing author, ForgeBots being configuration), and D1 is narrowed to the GitLab pusher. The new K5 query was checked against Ploeg's embedded PostgreSQL with all migrations and 7 seeded pull requests (3 clean of 7 as expected). The Loop dashboard panel in webgrip/homelab-cluster must be updated separately (filed on the Homelab Roadmap). Verified with `mise run verify` on the pinned toolchain (all gates passed). Ticket: https://vikunja.webgrip.dev/tasks/1750 🤖 Generated with [Claude Code](https://claude.com/claude-code)

ryangr0 added 1 commit

2026-10-03 11:50:33 +00:00

docs(kpis): count approvals as clean in K5 and drop the skip-review advice

[Workflow] On Pull Request / warnings (pull_request) Has been cancelled

Details

[Workflow] On Pull Request / release-policy (pull_request) Has been cancelled

Details

[Workflow] On Pull Request / checks (pull_request) Has been cancelled

Details

d9d0d7b0ee

K5 treated every submitted review as a change, so an approval lowered the
clean-merge rate, and the page told the owner to merge without a formal
approval to keep the proxy honest. That rewarded skipping review. Ploeg
already stores each review's verdict, reviewer, head SHA and time
(pull_request_reviews, migration 0023) and each push with its pusher
(pull_request_events, migration 0033).

K5 now counts a pull request as changed only when a review requested
changes, or someone other than the pull request's author pushed to the
branch, between the awaiting_review settle and the merge. Approvals and
comment-only reviews never lower it. The skip-review sentence is gone, the
data gap lists only what is still missing (GitLab pushes name no pusher,
the 500-event activity bound, a missing author, ForgeBots being
configuration), and D1 is narrowed to the GitLab pusher.

VIK-1750

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

ryangr0 merged commit ea7155c75e into development

2026-10-03 11:52:05 +00:00

ryangr0 referenced this pull request from a commit

2026-10-03 11:52:06 +00:00

Merge pull request 'docs(kpis): count approvals as clean in K5 and drop the skip-review advice' (#169) from ryangr0/kpi-k5-counts-approvals into development

Rows
Columns

docs(kpis): count approvals as clean in K5 and drop the skip-review advice #169