skills.rest
sebastianspicker
sebastianspickerCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

orchestration-ar

Launch parallel AI agent critiques on design.json and output review.json reports.

Updated Aug 23, 2026
Search Tags:#code quality#performance#security#orchestration#design review#adversarial review
One-click install
npx skills add https://github.com/sebastianspicker/phased-agent-orchestration --skill orchestration-ar
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: orchestration-ar
Source: https://github.com/sebastianspicker/phased-agent-orchestration/tree/main/.cursor/skills/orchestration-ar
Command: npx skills add https://github.com/sebastianspicker/phased-agent-orchestration --skill orchestration-ar

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the process of getting independent, multi-perspective critiques of a design artifact, ensuring thorough vetting before proceeding.

Core Features & Use Cases

  • Parallel Review Execution: Launches multiple specialized AI agents (architect, security, performance) to review a design concurrently.
  • Consolidated Findings: Aggregates and deduplicates findings from all reviewers.
  • Fact-Checking & Mitigation: Facilitates verification of findings against repository evidence and drafts mitigation recommendations.
  • Use Case: When a new design for a software feature is ready, this Skill can be invoked to have an architect, a security engineer, and a performance engineer independently review it, providing a comprehensive report of potential issues and their severity.

Quick Start

Use the orchestration-ar skill to perform an adversarial review on the current design artifact.

Frequently Asked Questions about orchestration-ar

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run an adversarial design review for a software feature?▼

To run an adversarial design review, provide a design.json file to launch parallel specialized AI agents that critique the artifact and output a consolidated review.json report.

What is parallel specialist critique for design artifacts?▼

Parallel specialist critique is an automated process that launches multiple AI agents, such as architect, security, and performance reviewers, to independently evaluate a design artifact concurrently.

Can I fact-check design review findings against repository evidence?▼

Yes, the adversarial review process facilitates fact-checking by verifying consolidated findings against repository evidence before drafting mitigation recommendations for human review.

Does the design review process output gate evaluations?▼

Yes, the design review process outputs gate evaluations by generating an adversarial-review-gate.json file alongside the consolidated review.json report to support decision-making.

What is the best way to consolidate multi-perspective critiques of a design?▼

The best way to consolidate multi-perspective critiques is to execute parallel agent reviews that aggregate and deduplicate findings into a single comprehensive report of potential issues and severities.

Do I need a specific input format to start an automated design critique?▼

Yes, you need to provide a specific design.json file as input to initialize the parallel adversarial review execution and generate the resulting review artifacts.

Related Skills

Explore More Software Engineering Skills→
zerAdazerAda

bmad-code-review

Automate adversarial code review with parallel review layers and structured triage.

Community
Advanced
mlopscommunitymlopscommunity
37

adversarial-code-review

Run three-agent adversarial code reviews on pull requests with confidence scoring.

Official
Advanced
ovargasovargas
12

verification-before-completion

Enforces a five-step gate workflow to verify completion claims with evidence.

Community
Advanced
iulianditaiuliandita
2

full-review

Orchestrate parallel execution of four audit skills for codebase quality assessment.

Community
Advanced