multi-review-opencode
/multi-review-opencode
Use when reviewing a PR, branch diff, or set of changed files with multiple parallel specialist perspectives (correctness, security, test coverage, and performance).
- Author
- gchiam ↗
- Repository
- claude-code-plugins ↗
- Licence
- MIT
- Last repository change
- 22.10.2026
Assessment and security
- Files in this package
- 4
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 22.10.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
Use when reviewing a PR, branch diff, or set of changed files with multiple parallel specialist perspectives (correctness, security, test coverage, and performance).
Original author description
Use when reviewing a PR, branch diff, or set of changed files with multiple parallel specialist perspectives (correctness, security, test coverage, and performance).
How it works
- Core Rules
- Execution Lifecycle
- References
SKILL.md preview
--- name: multi-review-opencode description: Use when reviewing a PR, branch diff, or set of changed files with multiple parallel specialist perspectives (correctness, security, test coverage, and performance). --- # OpenCode Multi-Review Run multi-perspective code reviews using parallel OpenCode subagents, independent validation passes, and aggregated summary reports. ## Core Rules 1. Scope: code review only. Never modify source files or apply automated fixes during review. 2. Never assume main is the base branch. Resolve base branch from PR metadata or git tracking. 3. Use background tasks for parallel review execution. 4. Normalize and validate findings before reporting. 5. Store artifacts in `.multi-reviews/`. ## Execution Lifecycle ### Phase 1: Target Resolution - Determine target: PR number, branch, or file list. - If target is a PR, fetch metadata using `gh pr view <number> --json baseRefName,title,body`. - Determine diff using verified base reference: `git diff origin/<baseRef>...HEAD`. - Triage changed files to identify applicable reviewer roles. ### Phase 2: Parallel Specialist Reviews - Select 2 to 3 relevant specialist roles from `references/reviewer-roles.md`. - Spawn each reviewer subagent in parallel via `task(..., run_in_background: true)`. - Use the contract from `references/reviewer-roles.md` to request structured output. - Wait for system notifications and collect results with `background_output`. ### Phase 3: Finding Validation - For any findings marked CRITICAL or IMPORTANT, spawn validation subagents in parallel to check for false positives against current file contents. - Downgrade or discard invalid findings based on validation results. ### Phase 4: Aggregation and Reporting - Deduplicate findings identified by multiple reviewers. - Generate `.multi-reviews/review-summary.md`. - Present a concise report table in the session, listing file, line, severity, and description. ## References - `references/reviewer-roles.md`: Specialist catalog and output JSON contract. - `references/workflow-steps.md`: Step-by-step dispatch and aggregation procedures. - `references/summary-template.md`: Report template for the final summary output.
Files in this package (4)
- SKILL.md 2 KB
- references/reviewer-roles.md 1 KB
- references/summary-template.md 1 KB
- references/workflow-steps.md 2 KB
Versions
The database has no other skills named multi-review-opencode.
Database snapshot: 13.11.2026 · package version 11c272d46e2d (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 95review
Multi-dimensional code review. Triggers on /codeforge:review or PR review. Correctness, security, performance, type safety, tests, standards. - 100code-review-checklist
Review checklist - correctness, security, performance, quality, tests, AI code. Use when reviewing code or a PR. - 90code-review
Multi-perspective code review covering readability, test coverage, performance, security, documentation accuracy, and best practices. Technology-agnostic — works on any language or stack. Use when reviewing a diff, PR, branch, or set of changed files before merge. - 100code-review-and-quality
Multi-axis code review (correctness, security, performance, design) before merging any change. Use to assess quality; not to refactor (use code-simplification) or hunt complexity only (use over-engineering-review). - 100code-review
Use when reviewing a pull request, diff, or branch. Produces severity-ranked findings covering correctness, security, performance, and maintainability, with concrete fixes rather than opinions. Trigger on "review this pr", "look over my diff before i push", "is this branch ready… - 100review-code
Multi-dimensional code review with structured reports. Analyzes correctness, readability, performance, security, testing, and architecture. Triggers on "review code", "code review", "审查代码", "代码审查".