plugins/compound-engineering/skills/resolve-pr-parallel/SKILL.md
Resolve all PR comments using parallel processing. Use when addressing PR review feedback, resolving review threads, or batch-fixing PR comments.
npx skillsauth add the-rabak/compound-engineering-plugin resolve-pr-parallelInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Resolve all unresolved PR review comments by spawning parallel agents for each thread.
Claude Code automatically detects git context:
Gather unresolved review comments. Use one of these methods:
Option A: User pastes comments Ask the user to paste the PR/MR review comments. This is the most reliable method and works with any Git platform.
Option B: GitHub CLI (gh)
bash ${CLAUDE_PLUGIN_ROOT}/skills/resolve-pr-parallel/scripts/get-pr-comments PR_NUMBER [OWNER/REPO]
Option C: GitLab CLI (glab)
glab mr view MR_NUMBER --comments
Option D: Platform API (if access token is available)
For GitHub:
gh api repos/OWNER/REPO/pulls/PR_NUMBER/reviews
gh api repos/OWNER/REPO/pulls/PR_NUMBER/comments
For GitLab:
curl --header "PRIVATE-TOKEN: $GITLAB_TOKEN" \
"https://YOUR_GITLAB_HOST/api/v4/projects/PROJECT_ID/merge_requests/MR_IID/discussions"
For Bitbucket:
curl -u "$BITBUCKET_USER:$BITBUCKET_TOKEN" \
"https://api.bitbucket.org/2.0/repositories/WORKSPACE/REPO/pullrequests/PR_ID/comments"
Create a TodoWrite list of all unresolved items grouped by type:
Spawn a pr-comment-resolver agent for each unresolved item in parallel.
Before dispatching pr-comment-resolver, use the platform's file-search tool against the bundled agent directory to look for pr-comment-resolver.md, then use the file-read tool to load the full template. Only if the bundled template cannot be loaded should you fall back to ov_load_global_agent "pr-comment-resolver". Before dispatching, quote the first non-empty line of the loaded template and record the source used. If you cannot quote the template because it was not found or could not be read, stop execution, raise the missing-template issue, and do not dispatch. Never dispatch a named agent by name alone.
If there are 3 comments, spawn 3 agents:
Always run all in parallel subagents/Tasks for each Todo item.
bash ${CLAUDE_PLUGIN_ROOT}/skills/resolve-pr-parallel/scripts/resolve-pr-thread THREAD_ID
Re-fetch comments to confirm all threads are resolved:
bash ${CLAUDE_PLUGIN_ROOT}/skills/resolve-pr-parallel/scripts/get-pr-comments PR_NUMBER
Should return an empty array []. If threads remain, repeat from step 1.
tools
Package one plan execution packet into a compact ticket-local execution packet with parent refs, scope fences, feature-home ownership, and evidence commands. Use when converting plans into local tickets or when execution needs one ticket-sized context pack without the full plan.
tools
Package one plan execution packet into a compact ticket-local execution packet with parent refs, scope fences, feature-home ownership, and evidence commands. Use when converting plans into local tickets or when execution needs one ticket-sized context pack without the full plan.
testing
Run a deep adversarial review of plans and architecture before implementation. Use when validating strategy docs, contracts, roadmaps, and competitive positioning with scored findings and prioritized recommendations.
testing
Run a deep adversarial review of plans and architecture before implementation. Use when validating strategy docs, contracts, roadmaps, and competitive positioning with scored findings and prioritized recommendations.