fix(skill): reviewer uses create_pull_review, not create_issue_comment #112
No reviewers
Labels
No labels
area:agents
area:dashboard
area:database
area:design
area:design-review
area:flows
area:infra
area:meta
area:security
area:sessions
area:webhook
area:workdir
security
type:bug
type:chore
type:meta
type:user-story
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
charles/claude-hooks!112
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "fix/reviewer-formal-pull-review"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Summary
The reviewer skill said "pick APPROVED or REQUEST_CHANGES — never COMMENT" but never named the tool. The agent interpreted that as "post a comment whose body begins with REQUEST_CHANGES", which renders fine in the UI but does not fire Forgejo's
pull_request_rejectedwebhook event — sohandleChangesRequestednever runs and the PR stalls silently.Hit live on PR #110 earlier today: reviewer posted change requests via
create_issue_comment, boss was never dispatched onaddress-review, and I had to manually escalate viacreate_pull_review(PR #110 issuecomment-6069) to unblock.Fix
skills/review.mdstep 6 now names the tool explicitly:mcp__forgejo__create_pull_reviewwithstate: "APPROVED" | "REQUEST_CHANGES".mcp__forgejo__create_issue_commentfor the verdict. Warning includes the #111 reference so the next reviewer reads the backstory if it forgets.body(markdown fine).Test plan
just qa— 357 pass, 0 failgrepconfirms the onlycreate_issue_commentreference underskills/is the warning line in review.mdlist_pull_reviewsshows astate: REQUEST_CHANGESentry and boss gets auto-dispatched onaddress-review(no manual escalation)Closes #111.
🤖 Generated with Claude Code