derek.holloway
7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
2.9 KiB
2.9 KiB
description, allowed-tools, argument-hint
| description | allowed-tools | argument-hint |
|---|---|---|
| Review a GitHub pull request using the AMD-SMI Review Agent | Bash(gh:*, git:*), Read, Write, Glob, Grep, WebFetch, Task | <PR_NUMBER> [review-type ...] — e.g. 1234 style tests (or just the number for comprehensive). Add 'fast' to skip rebuttal. Add 'no-build' to skip build step. |
Review a GitHub pull request using the AMD-SMI Review Agent.
Arguments
$ARGUMENTScontains:<PR_NUMBER> [review-type ...]- First argument: PR number (required). Construct the full URL as
https://github.com/ROCm/rocm-systems/pull/<PR_NUMBER> - Remaining: optional review types:
style,tests,docs,architecture,security,performance,build,skeptic - Special modifier:
fast— skips the rebuttal round (comprehensive mode only) - Special modifier:
no-build— skips the build & install step - If no types specified, perform a comprehensive review (all subagents, with rebuttal)
Process
1. Construct PR URL
Set PR_URL=https://github.com/ROCm/rocm-systems/pull/<PR_NUMBER> from the first argument.
2. Fetch PR Information
gh pr view $PR_URL --json number,title,author,body,files,additions,deletions,state,baseRefName,headRefName,comments,reviews
3. Fetch the Diff
gh pr diff $PR_URL
4. Gather CI Evidence
Check for linked CI runs and fetch step-level data:
# Get CI check runs for the PR
gh pr checks $PR_URL
# For each failed or interesting run, get details
gh run view <RUN_ID> --json jobs
Find a recent baseline run on main for comparison:
gh run list --branch main --workflow <WORKFLOW> --limit 1 --json databaseId,conclusion
Compare step timings and status between PR run and baseline. Pass this evidence to the test and performance subagents.
5. Fetch Unresolved Comments
gh pr view $PR_URL --json comments,reviews,reviewRequests
Extract unresolved review comments for the unresolved comments analysis.
6. Dispatch Review
Invoke the AMD-SMI Review Agent with:
- PR metadata (title, author, files, +/- lines)
- The diff
- The review type(s) from
$ARGUMENTS(or "comprehensive" if none) - CI evidence (test results, step timings, baseline comparison)
- Unresolved comments
The agent will dispatch to the appropriate subagent(s) and produce a formatted review.
By default, comprehensive reviews include a rebuttal round (Round 2) with the skeptic subagent. If fast was specified, the rebuttal round is skipped.
7. Report Results
- Present the review inline
- Summarize the overall assessment
- List any blocking issues found
- If saving requested, use:
reviews/pr_{NUMBER}[_{TYPE}].md
Examples
/amdsmi-review-pr https://github.com/ROCm/rocm-systems/pull/123
/amdsmi-review-pr https://github.com/ROCm/rocm-systems/pull/123 style
/amdsmi-review-pr https://github.com/ROCm/rocm-systems/pull/123 tests security performance
/amdsmi-review-pr https://github.com/ROCm/rocm-systems/pull/123 fast