fix: responder uses gh api graphql to read review threads#147
Merged
Conversation
There was a problem hiding this comment.
Pull request overview
Updates the Review Responder workflow instructions to use gh api graphql for fetching unresolved review threads (instead of MCP tooling), and adjusts the label/processing sequence to avoid adding loop-guard labels when there’s nothing to address.
Changes:
- Replaces the “check review comments / threads” step with a GraphQL-based
gh api graphqlquery and JSON filtering instructions. - Reorders the loop-guard label logic so
review-response-attemptedis only applied after confirming unresolved threads exist. - Updates the per-thread handling instructions to rely on the GraphQL thread data (path/line/body).
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
microsasa
force-pushed
the
fix/responder-graphql-threads
branch
2 times, most recently
from
1052fdf to
5a14e5c
Compare
There was a problem hiding this comment.
Pull request overview
Updates the Review Responder gh-aw workflow instructions to change when the workflow stops, how it guards against loops, and how it sequences label application vs. processing review threads.
Changes:
- Simplifies the “no-comment review” handling by stopping immediately when the triggering review has no inline comments.
- Reorders loop-guard labeling (
review-response-attempted) to occur earlier in the process. - Adjusts the unresolved-thread handling instructions (including the “>10 threads” guidance) and trims push/reply guidance.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Reverts the review-responder to the original simple instructions that successfully addressed review threads (run 23117978475, 2026-03-15). Multiple rounds of 'improvements' broke the responder's ability to read threads — over-specifying how to query threads (GraphQL, MCP) instead of letting the agent figure it out, and changing step ordering that prevented thread discovery. Only change from the known-working version: removed resolve-thread safe-output (now handled by the pipeline orchestrator) and kept the if: label gate (saves compute on non-aw PRs). Closes #148 Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
microsasa
force-pushed
the
fix/responder-graphql-threads
branch
from
5a14e5c to
fd42698
Compare
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
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.
Reverts the review-responder to the original simple instructions that successfully addressed review threads (run
23117978475, 2026-03-15). Multiple rounds of CLI-driven churn broke the responder's ability to read threads by over-specifying how to query them.Only changes from the known-working version:
resolve-pull-request-review-threadsafe-output (orchestrator handles this now)if:label gate (saves compute on non-aw PRs)Closes #148