How long does the integration take?

Change one base URL in an OpenAI-compatible client, or install the Python or TypeScript SDK; the quickstart routes a first request in a few minutes.

Will this lock us into another vendor?

No. The router is Apache-2.0 open source, your provider keys stay yours, and the API is OpenAI-compatible, so leaving is a change of base URL.

What happens when a provider fails?

The router falls back to the next model your policy allows and records the fallback in the trace, so an outage costs a retry, not an incident.

conduct_codereview - Prompt

Name: conduct_codereview
Author: focusconsulting

Prompt file imported from focusconsulting/pa-compact-licensing-platform (.claude/commands/conduct_codereview.md). Copyright stays with the author.

Perform a comprehensive code quality review of the specified file(s) or recent changes using specialized sub-agents.

Review Process

Launch 5 specialized sub-agents in parallel to analyze the code from different perspectives:

Factual Reviewer (subagent_type: general-purpose)
- Verify technical accuracy against official documentation
- Check API usage matches documentation
- Validate framework patterns and best practices
- Confirm dependency versions and compatibility
Senior Engineer (subagent_type: general-purpose)
- Review architecture decisions and design patterns
- Evaluate code structure and organization
- Assess maintainability and scalability
- Check for proper separation of concerns
- Identify potential technical debt
Security Expert (subagent_type: general-purpose)
- Identify security vulnerabilities (OWASP Top 10)
- Check for injection vulnerabilities (SQL, XSS, command injection)
- Review authentication and authorization logic
- Assess data validation and sanitization
- Check for exposed secrets or credentials
- Review error handling for information leakage
Consistency Reviewer (subagent_type: general-purpose)
- Check adherence to project coding standards
- Verify naming conventions consistency
- Review code formatting and style
- Check import organization and structure
- Validate TypeScript/Python type usage
- Ensure consistent error handling patterns
Redundancy Checker (subagent_type: general-purpose)
- Identify duplicate code and logic
- Find opportunities for abstraction
- Locate similar functions that could be consolidated
- Identify repeated patterns that could be utilities
- Check for unnecessary code complexity

Usage

When invoked, you should:

Ask the user what to review:
- Specific file(s) path(s)
- Recent git changes (staged or committed)
- Entire module or directory
- Current IDE selection
Launch all 5 agents in parallel using a single message with multiple Task tool calls
Wait for all agents to complete
Synthesize findings into a comprehensive report with:
- Executive Summary (critical issues first)
- Findings by category (Factual, Architecture, Security, Consistency, Redundancy)
- Prioritized action items (Critical → High → Medium → Low)
- Code examples and suggested fixes
- Positive observations (what's done well)

Report Format

# Code Quality Review Report

## Executive Summary
[High-level overview with critical issues highlighted]

## Critical Issues 🚨
[Issues requiring immediate attention]

## Findings by Reviewer

### 📋 Factual Accuracy
- [Findings from factual reviewer]

### 🏗️ Architecture & Design
- [Findings from senior engineer]

### 🔒 Security
- [Findings from security expert]

### 📏 Consistency & Standards
- [Findings from consistency reviewer]

### ♻️ Code Redundancy
- [Findings from redundancy checker]

## Action Items

### Priority 1 (Critical)
- [ ] [Action item with file references]

### Priority 2 (High)
- [ ] [Action item with file references]

### Priority 3 (Medium)
- [ ] [Action item with file references]

### Priority 4 (Low/Enhancement)
- [ ] [Action item with file references]

## Positive Observations ✨
[Things done well that should be maintained]

## Metrics
- Files reviewed: X
- Issues found: X (Critical: X, High: X, Medium: X, Low: X)
- Code redundancy: X instances
- Security concerns: X

Context-Specific Guidelines

For client/ (Next.js / TypeScript)

Check Next.js App Router patterns and conventions
Review React component structure and hooks usage
Validate TypeScript types and interfaces
Check client/server component boundaries
Review styling patterns (PostCSS)
Validate i18n implementation

For api/ (Python / FastAPI)

Check FastAPI route patterns and dependencies
Review Pydantic model usage and validation
Validate database access patterns
Check SQL injection prevention
Review authentication and authorization logic
Validate error handling and HTTP status codes

For iac/ (Infrastructure as Code)

Check Terraform resource naming conventions
Review security group and IAM policies
Validate variable usage and defaults
Check for hardcoded values that should be variables

Important Notes

Run all agents in parallel for efficiency
Each agent should focus on their specialty
Agents should read relevant files and context
Provide specific file paths and line numbers in findings
Include code snippets showing issues and fixes
Prioritize findings by impact and effort

Install conduct_codereview and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

Perform a comprehensive code quality review of the specified file(s) or recent changes using specialized sub-agents. ## Review Process Launch 5 specialized sub-agents in parallel to analyze the code from different perspectives: 1. **Factual Reviewer** (subagent_type: general-purpose) - Verify technical accuracy against official documentation - Check API usage matches documentation - Validate framework patterns and best practices - Confirm dependency versions and compatibility 2. **Senior Engineer** (subagent_type: general-purpose) - Review architecture decisions and design patterns - Evaluate code structure and organization - Assess maintainability and scalability - Check for proper separation of concerns - Identify potential technical debt 3. **Security Expert** (subagent_type: general-purpose) - Identify security vulnerabilities (OWASP Top 10) - Check for injection vulnerabilities (SQL, XSS, command injection) - Review authentication and authorization logic - Assess data validation and sanitization - Check for exposed secrets or credentials - Review error handling for information leakage 4. **Consistency Reviewer** (subagent_type: general-purpose) - Check adherence to project coding standards - Verify naming conventions consistency - Review code formatting and style - Check import organization and structure - Validate TypeScript/Python type usage - Ensure consistent error handling patterns 5. **Redundancy Checker** (subagent_type: general-purpose) - Identify duplicate code and logic - Find opportunities for abstraction - Locate similar functions that could be consolidated - Identify repeated patterns that could be utilities - Check for unnecessary code complexity ## Usage When invoked, you should: 1. Ask the user what to review: - Specific file(s) path(s) - Recent git changes (staged or committed) - Entire module or directory - Current IDE selection 2. Launch all 5 agents in parallel using a single message with multiple Task tool calls 3. Wait for all agents to complete 4. Synthesize findings into a comprehensive report with: - Executive Summary (critical issues first) - Findings by category (Factual, Architecture, Security, Consistency, Redundancy) - Prioritized action items (Critical → High → Medium → Low) - Code examples and suggested fixes - Positive observations (what's done well) ## Report Format ```markdown # Code Quality Review Report ## Executive Summary [High-level overview with critical issues highlighted] ## Critical Issues 🚨 [Issues requiring immediate attention] ## Findings by Reviewer ### 📋 Factual Accuracy - [Findings from factual reviewer] ### 🏗️ Architecture & Design - [Findings from senior engineer] ### 🔒 Security - [Findings from security expert] ### 📏 Consistency & Standards - [Findings from consistency reviewer] ### ♻️ Code Redundancy - [Findings from redundancy checker] ## Action Items ### Priority 1 (Critical) - [ ] [Action item with file references] ### Priority 2 (High) - [ ] [Action item with file references] ### Priority 3 (Medium) - [ ] [Action item with file references] ### Priority 4 (Low/Enhancement) - [ ] [Action item with file references] ## Positive Observations ✨ [Things done well that should be maintained] ## Metrics - Files reviewed: X - Issues found: X (Critical: X, High: X, Medium: X, Low: X) - Code redundancy: X instances - Security concerns: X ``` ## Context-Specific Guidelines ### For client/ (Next.js / TypeScript) - Check Next.js App Router patterns and conventions - Review React component structure and hooks usage - Validate TypeScript types and interfaces - Check client/server component boundaries - Review styling patterns (PostCSS) - Validate i18n implementation ### For api/ (Python / FastAPI) - Check FastAPI route patterns and dependencies - Review Pydantic model usage and validation - Validate database access patterns - Check SQL injection prevention - Review authentication and authorization logic - Validate error handling and HTTP status codes ### For iac/ (Infrastructure as Code) - Check Terraform resource naming conventions - Review security group and IAM policies - Validate variable usage and defaults - Check for hardcoded values that should be variables ## Important Notes - Run all agents in parallel for efficiency - Each agent should focus on their specialty - Agents should read relevant files and context - Provide specific file paths and line numbers in findings - Include code snippets showing issues and fixes - Prioritize findings by impact and effort

{ "prompt": "Perform a comprehensive code quality review of the specified file(s) or recent changes using specialized sub-agents.\n\n## Review Process\n\nLaunch 5 specialized sub-agents in parallel to analyze the code from different perspectives:\n\n1. **Factual Reviewer** (subagent_type: general-purpose)\n - Verify technical accuracy against official documentation\n - Check API usage matches documentation\n - Validate framework patterns and best practices\n - Confirm dependency versions and compatibility\n\n2. **Senior Engineer** (subagent_type: general-purpose)\n - Review architecture decisions and design patterns\n - Evaluate code structure and organization\n - Assess maintainability and scalability\n - Check for proper separation of concerns\n - Identify potential technical debt\n\n3. **Security Expert** (subagent_type: general-purpose)\n - Identify security vulnerabilities (OWASP Top 10)\n - Check for injection vulnerabilities (SQL, XSS, command injection)\n - Review authentication and authorization logic\n - Assess data validation and sanitization\n - Check for exposed secrets or credentials\n - Review error handling for information leakage\n\n4. **Consistency Reviewer** (subagent_type: general-purpose)\n - Check adherence to project coding standards\n - Verify naming conventions consistency\n - Review code formatting and style\n - Check import organization and structure\n - Validate TypeScript/Python type usage\n - Ensure consistent error handling patterns\n\n5. **Redundancy Checker** (subagent_type: general-purpose)\n - Identify duplicate code and logic\n - Find opportunities for abstraction\n - Locate similar functions that could be consolidated\n - Identify repeated patterns that could be utilities\n - Check for unnecessary code complexity\n\n## Usage\n\nWhen invoked, you should:\n\n1. Ask the user what to review:\n - Specific file(s) path(s)\n - Recent git changes (staged or committed)\n - Entire module or directory\n - Current IDE selection\n\n2. Launch all 5 agents in parallel using a single message with multiple Task tool calls\n\n3. Wait for all agents to complete\n\n4. Synthesize findings into a comprehensive report with:\n - Executive Summary (critical issues first)\n - Findings by category (Factual, Architecture, Security, Consistency, Redundancy)\n - Prioritized action items (Critical → High → Medium → Low)\n - Code examples and suggested fixes\n - Positive observations (what's done well)\n\n## Report Format\n\n```markdown\n# Code Quality Review Report\n\n## Executive Summary\n[High-level overview with critical issues highlighted]\n\n## Critical Issues 🚨\n[Issues requiring immediate attention]\n\n## Findings by Reviewer\n\n### 📋 Factual Accuracy\n- [Findings from factual reviewer]\n\n### 🏗️ Architecture & Design\n- [Findings from senior engineer]\n\n### 🔒 Security\n- [Findings from security expert]\n\n### 📏 Consistency & Standards\n- [Findings from consistency reviewer]\n\n### ♻️ Code Redundancy\n- [Findings from redundancy checker]\n\n## Action Items\n\n### Priority 1 (Critical)\n- [ ] [Action item with file references]\n\n### Priority 2 (High)\n- [ ] [Action item with file references]\n\n### Priority 3 (Medium)\n- [ ] [Action item with file references]\n\n### Priority 4 (Low/Enhancement)\n- [ ] [Action item with file references]\n\n## Positive Observations ✨\n[Things done well that should be maintained]\n\n## Metrics\n- Files reviewed: X\n- Issues found: X (Critical: X, High: X, Medium: X, Low: X)\n- Code redundancy: X instances\n- Security concerns: X\n```\n\n## Context-Specific Guidelines\n\n### For client/ (Next.js / TypeScript)\n- Check Next.js App Router patterns and conventions\n- Review React component structure and hooks usage\n- Validate TypeScript types and interfaces\n- Check client/server component boundaries\n- Review styling patterns (PostCSS)\n- Validate i18n implementation\n\n### For api/ (Python / FastAPI)\n- Check FastAPI route patterns and dependencies\n- Review Pydantic model usage and validation\n- Validate database access patterns\n- Check SQL injection prevention\n- Review authentication and authorization logic\n- Validate error handling and HTTP status codes\n\n### For iac/ (Infrastructure as Code)\n- Check Terraform resource naming conventions\n- Review security group and IAM policies\n- Validate variable usage and defaults\n- Check for hardcoded values that should be variables\n\n## Important Notes\n\n- Run all agents in parallel for efficiency\n- Each agent should focus on their specialty\n- Agents should read relevant files and context\n- Provide specific file paths and line numbers in findings\n- Include code snippets showing issues and fixes\n- Prioritize findings by impact and effort", "variables": [], "notes": "Perform a comprehensive code quality review of the specified file(s) or recent changes using specialized sub-agents.", "metadata": { "source": { "provider": "github-claude-commands", "repository": "https://github.com/focusconsulting/pa-compact-licensing-platform", "path": ".claude/commands/conduct_codereview.md", "ref": "69e0bb30570f6bfa1f4c89059497916ac52d5bcb", "url": "https://github.com/focusconsulting/pa-compact-licensing-platform/blob/69e0bb30570f6bfa1f4c89059497916ac52d5bcb/.claude/commands/conduct_codereview.md", "key": "focusconsulting/pa-compact-licensing-platform/.claude/commands/conduct_codereview.md" } } }

conduct_codereview

About

Review Process

Usage

Report Format

Context-Specific Guidelines

For client/ (Next.js / TypeScript)

For api/ (Python / FastAPI)

For iac/ (Infrastructure as Code)

Important Notes

Install conduct_codereview and it becomes one more candidate the router can pick - when it fits.

What it costs to adopt - in engineering time.

One base URL

No vendor lock-in

Failover built in

Use it

Manifest

Reviews