schlenks/multi-review-aggregation
plugins/superpowers-bd/skills/multi-review-aggregation/SKILL.md
Use when dispatching code reviews for tiers with N greater than 1 (max-20x, max-5x) in subagent-driven development, or manually for critical changes over 200 lines or security-sensitive code
$ install --global
skillsauthnpx skillsauth add schlenks/superpowers-bd multi-review-aggregationInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: May 10, 2026, 4:20 AM213.3s6 files scanned
SKILL.md
- name:
- multi-review-aggregation
- description:
- Use when dispatching code reviews for tiers with N greater than 1 (max-20x, max-5x) in subagent-driven development, or manually for critical changes over 200 lines or security-sensitive code
- effort:
- xhigh
Multi-Review Aggregation
Dispatch N independent code reviews and aggregate findings. Each reviewer catches different bugs -- union preserves the long tail that single-shot misses.
Research basis: SWR-Bench (arXiv 2509.01494) -- N independent reviews: 43.67% F1 improvement, 118.83% recall improvement. Diminishing returns past N=5; N=3 captures most improvement.
Core principle: Independence via separate Task dispatches -- same base prompt, no shared context.
N Selection by Tier
| Tier | N Reviews | Rationale | |------|-----------|-----------| | max-20x | 3 | Quality priority -- full aggregation | | max-5x | 3 | Balanced -- same recall benefit | | pro/api | 1 | Budget priority -- single review |
When N=1, skip this skill -- use standard single code review.
Parallel Dispatch Pattern
After spec review passes, dispatch N independent reviews (run_in_background=True, each gets "Reviewer i of N"). Always aggregate when 2+ reviewers succeed — even unanimous approval may contain different Minor findings, Suggestions, or Not Checked items. Dispatch aggregator (haiku model).
Full dispatch code: see references/dispatch-code.md. Aggregator prompt: see ./aggregator-prompt.md.
Aggregation Algorithm
Severity Voting
| Condition | Result | |-----------|--------| | All reviewers agree on severity | Keep that severity | | Reviewers disagree | Use highest severity | | Lone finding Critical or Important | Keep original severity (no downgrade) | | Lone finding Minor | Downgrade to Suggestion | | Lone finding BUT security or data-loss | Keep original severity (no downgrade) |
Severity levels: Critical > Important > Minor > Suggestion
Verdict
- "Ready to merge: Yes" -- zero Critical AND zero Important AND majority approved
- "Ready to merge: With fixes" -- only Minor/Suggestion after aggregation
- "Ready to merge: No" -- any Critical or Important remain
Full deduplication/merging rules: see references/aggregation-details.md.
Output Format
## Strengths
- [strength] [Reviewers: 1, 2, 3]
## Issues
### Critical / Important / Minor / Suggestion
- [issue] [Reviewers: N, N] -- file:line
(note downgrade/security provenance as applicable)
## Uncovered Paths
- [path/scenario] [Reviewers: X, Y]
## Not Checked
- [area] [Reviewers: X, Y]
## Assessment
Ready to merge: [Yes/With fixes/No]
Reviewers: X/N approved, Y requested changes
Full format spec: see references/output-provenance.md.
Red Flags
Never:
- Share context between reviewers (defeats independence)
- Use N>1 for pro/api tier (budget constraint)
- Skip aggregation when reviewers disagree
- Downgrade security findings even as lone findings
Always:
- Dispatch all N reviews in parallel
- Include reviewer number in each dispatch prompt
- Use haiku for aggregation
- Record per-reviewer metrics separately
Reference Files
references/dispatch-code.md: Full dispatch flow with on_spec_review_pass handlerreferences/aggregation-details.md: Deduplication, strengths merging, malformed output, timeout recoveryreferences/output-provenance.md: Provenance annotation rules and full output format specreferences/metrics-and-cost.md: Per-reviewer metric keys, cost impact, per-tier breakdownaggregator-prompt.md: Aggregator Task dispatch prompt template
Related Skills
schlenks/plan2beads
tools
VerifiedTrustedCommunity
Use when converting a Superpowers-BD implementation plan or Shortcut story into a beads epic with dependency-aware child tasks
2SKILL.mdUpdated May 16, 2026
schlenks/ad-hoc-code-review
development
VerifiedTrustedCommunity
Use when the user asks for /cr-style review of local changes, commits, a branch diff, or a GitHub PR outside subagent-driven development
2SKILL.mdUpdated May 16, 2026
schlenks/writing-plans
development
VerifiedTrustedCommunity
Use when you have a spec or requirements for a multi-step task, before touching code
2SKILL.mdUpdated May 16, 2026
schlenks/using-superpowers
tools
VerifiedTrustedCommunity
Use when starting any conversation - establishes how to find and use skills, requiring Skill tool invocation before ANY response including clarifying questions
2SKILL.mdUpdated May 16, 2026