$ loading_
Review system designs from six critical perspectives and produce unified risk assessments.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "adversarial-review" skill from askskill: 1. Download https://raw.githubusercontent.com/microsoft/amplifier-bundle-systems-design/main/skills/adversarial-review/SKILL.md 2. Save it as ~/.claude/skills/adversarial-review/SKILL.md 3. Reload skills and tell me it's ready
Please perform an adversarial review of this payment system design from six perspectives: SRE, security, staff engineer, finance, operator, and developer advocate. The design includes multi-region deployment, a Kafka event bus, PostgreSQL primary-replica setup, Redis cache, and a third-party fraud detection API. Output the major risks, severity, trigger conditions, and a unified prioritization plan.
A risk list broken down by six perspectives, plus consolidated high-priority remediation recommendations.
I’m reviewing a design for an internal logging platform. Act as six tough reviewers and raise the sharpest questions about availability, permission isolation, scalability, cost, operational complexity, and developer onboarding experience, then provide a unified risk conclusion.
A set of sharp discussion prompts suitable for live review meetings, along with a final consolidated risk judgment.
Review this AI inference service design and focus on overlooked risks from finance, SRE, and security perspectives: GPU autoscaling, usage-based billing, 180-day log retention, fallback to external APIs, and a public SDK for third parties. End with a unified risk ranking and improvement recommendations.
A combined analysis of hidden cost, reliability, and security issues with actionable optimization recommendations.
You are conducting an adversarial review of a system design. Your job is to find flaws, not to praise. You are a critic, not a collaborator.
The design to review is provided via $ARGUMENTS. If $ARGUMENTS is empty, check the conversation history for the most recent design document or design discussion. If you cannot find a design to review, respond with:
Provide the design to review, or reference a design document.
Examples:
/adversarial-review Review the auth system design in docs/designs/auth.md
/adversarial-review [paste or describe the design]
Before launching reviewers, read and understand the design:
Use the delegate tool to launch all six agents concurrently in a single message. Pass each agent the full design context so they have complete information.
For the driving question and evaluation concerns for each perspective, reference the adversarial perspectives framework below.
Delegate with instruction: "You are a senior SRE. Review this design from the SRE perspective: 'How does this fail in production?' [full design context]"
Delegate with instruction: "You are a security reviewer. Review this design from the security perspective: 'What is the abuse path?' [full design context]"
Delegate with instruction: "You are a staff engineer. Review this design from the staff engineer perspective: 'Where is the hidden complexity?' [full design context]"
Delegate with instruction: "You are a finance/cost owner. Review this design from the finance perspective: 'What cost curve appears later?' [full design context]"
Delegate with instruction: "You are the person who operates this at 2am. Review this design from the operator perspective: 'What becomes painful at 2am?' [full design context]"
Delegate with instruction: "You are a developer advocate who cares about developer experience. Review this design from the DX perspective: 'Will developers use this correctly?' Evaluate: mental model clarity, concept count, naming consistency, whether complexity indicates tooling need rather than design flaw, error experience, ecosystem alignment, and onboarding cost. [full design context]"
Wait for all six agents to complete. Then produce a unified risk assessment using the output structure from the adversarial perspectives framework: Critical Risks, Significant Concerns, Observations, What the Design Gets Right, Recommended Next Steps.
@systems-design:context/adversarial-perspectives.md
Design and assess enterprise integration patterns, legacy modernization, and orchestration strategies.
Review system designs with a seven-step method to surface risks and improvements.
Evaluate system designs through Unix/Linux principles for simplicity and composability.
Design or assess Azure system architectures, operations, and cloud service choices.
Design and evaluate data pipelines, streaming systems, and ETL architecture patterns.
Design offline-first edge systems with sync, conflict handling, and weak-network resilience.
Review code deeply across correctness, tests, security, performance, and product quality.
Scan LLM apps for security flaws and return prioritized vulnerability findings.
Run multi-model reviews for code changes, PRs, and risky edits.
Refine product specifications through multi-model debate until consensus is reached.
Review code or branches for correctness, compatibility, architecture, tests, performance, and security.
Generate or audit design systems and review styling consistency changes.