A comprehensive collection of 24+ expert agents (including FinTech & Healthcare specialists), 12 custom tools with built-in analytics and resource discovery, 5 language instructions, 5 prompts, and 5 comprehensive skills for GitHub Copilot, now powered by Specification-Driven Development (SDD).
Agent Pro provides 40+ curated resources plus SDD methodology framework to enhance your development workflow with AI-powered assistance across multiple domains including architecture, security, testing, cloud infrastructure, regulated industries (finance, healthcare), and systematic feature development. All resources are interconnected via the Related Resources feature.
- 24 Expert Agents - Specialized AI assistants accessible via
@in Copilot Chat - 12 Custom Tools - Code analysis, test generation, documentation, performance profiling, dependency analysis, API design, 5 SpecKit SDD tools, plus resourceDiscovery
- Specification-Driven Development - Complete SDD framework with constitutional governance, templates, and validation
- Built-in Analytics - Track tool usage, success rates, and performance (privacy-respecting, local only)
- Copilot Integration - All agents appear natively in GitHub Copilot's @ menu
- 5 Language Instructions - Auto-applied coding standards for Python, TypeScript, Go, Rust, Java
- 5 Reusable Prompts - Pre-built templates for common tasks
- 5 Comprehensive Skills - Bundled capabilities with best practices
- Constitutional Framework - 8 architectural principles governing all development
- Enhanced Validation - Production-ready validation with framework version checks and security pattern detection
- Comprehensive Testing - 53 functional tests ensure production quality
- Global Storage - Agents available in all your projects
- Install Agent Pro from the VS Code Marketplace
- Open GitHub Copilot Chat (
Ctrl+Alt+I/Cmd+Option+I) - Type
@to see all 24 expert agents - Select an agent (e.g.,
@PythonExpertor@FinTechExpert) and start chatting! - Agents automatically have access to 6 custom tools for enhanced capabilities
- View usage statistics:
Ctrl+Shift+P→ "Agent Pro: Show Usage Statistics"
Open GitHub Copilot Chat and type @ to see all available agents. All agents now have access to these custom tools:
- codeAnalyzer - Analyzes code complexity, metrics, and patterns
- testGenerator - Suggests test strategies for your code
- documentationBuilder - Generates documentation templates
- performanceProfiler - Detects performance anti-patterns
- dependencyAnalyzer - Scans project dependencies (package.json, requirements.txt, go.mod, Cargo.toml)
- apiDesigner - Provides OpenAPI 3.1 specifications and REST API best practices
Example:
@python-expert analyze this code for complexity
@testing-specialist suggest tests for this function
@performance-expert profile this algorithm
@fintech-expert review this payment processing code for PCI DSS compliance
@healthcare-expert validate this FHIR resource
| Agent | Specialization | Use When |
|---|---|---|
| @accessibility-expert | WCAG 2.2, ARIA, inclusive design | Building accessible UIs |
| @ai-ml-engineering-expert | Agentic AI, MLOps, PyTorch, RAG | ML/AI development |
| @architecture-expert | Clean Architecture, DDD, system design | Designing systems |
| @cloud-architect | AWS, Azure, GCP, multi-cloud | Cloud infrastructure |
| @code-reviewer | SOLID, security, maintainability | Code reviews |
| @data-engineering-expert | dbt, Airflow, Snowflake, Data Mesh | Data pipelines |
| @design-patterns-expert | SOLID, GoF 23 patterns | Design patterns |
| @design-systems-expert | Figma Variables, Design Tokens | Design systems |
| @devops-expert | Docker, Kubernetes, CI/CD | DevOps workflows |
| @documentation-expert | Technical writing, API docs | Documentation |
| @fintech-expert | PCI DSS, payments, fraud, KYC/AML | Financial applications |
| @functional-programming-expert | FRP, RxJS, monads | Functional code |
| @graphql-expert | Apollo Federation 2.10 | GraphQL APIs |
| @healthcare-expert | HIPAA, HL7/FHIR, EHR integration | Healthcare applications |
| @microservices-expert | SAGA, Circuit Breaker, CQRS | Microservices |
| @mobile-development-expert | React Native, Flutter, iOS, Android | Mobile apps |
| @observability-sre-expert | OpenTelemetry, SLOs | Monitoring & SRE |
| @performance-expert | Profiling, optimization | Performance tuning |
| @platform-engineering-expert | Backstage IDP, Golden Paths | Platform engineering |
| @python-expert | Python 3.12+, asyncio, type hints | Python development |
| @security-expert | OWASP Top 10, secure coding | Security reviews |
| @systems-programming-expert | Rust, WebAssembly | Systems programming |
| @testing-specialist | TDD/BDD, quality assurance | Testing strategies |
| @typescript-expert | TypeScript 5.x, React, Node.js | TypeScript/JavaScript |
Prompts are reusable templates that work with specific agents:
- Code Review - Comprehensive code review checklist
- Conventional Commit - Generate semantic commit messages
- Generate README - Create comprehensive README files
- Generate Tests - Generate test suites (unit/integration/E2E)
- Refactor Code - Refactor for quality and maintainability
Instructions automatically apply to files matching patterns:
- Python (
**/*.py) - PEP 8, type hints, asyncio, modern best practices - TypeScript (
**/*.ts,**/*.tsx) - TS 5.x standards, React patterns - Go (
**/*.go) - Effective Go, idioms, error handling - Rust (
**/*.rs) - Rust API Guidelines, ownership, safety patterns - Java (
**/*.java) - Modern Java 17+, records, sealed classes, streams
Skills are comprehensive, self-contained capabilities:
- API Development - RESTful APIs, FastAPI, Express, OpenAPI
- Database Design - PostgreSQL, schemas, normalization, optimization
- GraphQL & Microservices - Federation 2.10, distributed patterns
- Multi-Agent Orchestration - Agent HQ, workflow coordination
- Testing Strategies - TDD/BDD, shift-left/right, test pyramid
All agents have access to 7 specialized tools (plus 5 SpecKit tools):
Analyzes code complexity, metrics, and patterns:
- Total lines, code lines, comment lines
- Average line length and comment ratio
- Language-specific insights
Suggests test strategies and frameworks:
- Recommends appropriate testing framework (Jest, pytest, JUnit, etc.)
- Identifies test types needed (unit, integration, edge cases)
- Analyzes code structure for testability
Generates documentation templates:
- Language-specific doc formats (JSDoc, TSDoc, docstring, JavaDoc, etc.)
- Function/method documentation structure
- Usage examples and best practices
Detects performance anti-patterns:
- Array operations in loops
- Inefficient chaining (map+filter)
- Deep clone anti-patterns
- Framework-specific optimizations
Scans and analyzes project dependencies:
- Supports: package.json (Node.js), requirements.txt (Python), go.mod (Go), Cargo.toml (Rust)
- Detects outdated packages and provides update recommendations
- Security scanning suggestions (Dependabot, Renovate, pip-audit, cargo audit)
REST API design and OpenAPI specification generator:
- OpenAPI 3.1 templates and best practices
- HTTP verb mapping and resource naming conventions
- API design patterns (pagination, filtering, sorting, HATEOAS)
- Security guidelines (OAuth 2.0, JWT, rate limiting)
- Framework-specific tool recommendations (FastAPI, Express, SpringDoc)
Discover all available Agent Pro resources:
- Lists all agents, prompts, skills, instructions, and templates
- Helps find the right resource for your task
- Shows relationships between resources
1. @graphql-expert design my API schema
2. Use codeAnalyzer tool to check complexity
3. @security-expert review authentication
4. @testing-specialist create integration tests (with testGenerator)
5. @documentation-expert generate API docs (with documentationBuilder)
1. @mobile-development-expert help me structure this React Native app
2. @design-systems-expert create component library
3. @performance-expert optimize rendering (with performanceProfiler)
4. @accessibility-expert ensure WCAG compliance
1. @data-engineering-expert design ETL pipeline
2. @cloud-architect set up infrastructure
3. @observability-sre-expert add monitoring
4. @testing-specialist create data quality tests
1. @ai-ml-engineering-expert help build RAG system
2. @python-expert optimize model code (with codeAnalyzer)
3. @platform-engineering-expert create deployment pipeline
4. @observability-sre-expert add model monitoring
Key agents now have critical content embedded inline for consistent, self-contained operation:
- TypeScript Expert: Embedded strict tsconfig requirements, naming conventions, anti-patterns to flag
- Python Expert: Embedded PEP 8 rules, type hint requirements, import ordering
- Code Reviewer: Embedded security/quality/performance checklists, severity classification
- Testing Specialist: Embedded TDD methodology, testing trophy, coverage targets
- DevOps Expert: Embedded conventional commits format, commit types, CI/CD examples
- Security Expert: Embedded OWASP Top 10 checklist, security headers, critical checks
- New agent-quickref.md - Quick reference for all 24 agents
- Shows embedded resources summary per agent
- Custom tools reference table
- Quick commands cheatsheet
Agent Pro: Show Agent Quick Reference- View embedded resources for key agents via quick pick
- 53 functional tests (up from 45)
- Tests verify embedded content in all enhanced agents
- Complete SDD methodology framework
- Constitutional governance with 8 architectural principles
- 4 SDD templates: spec, plan, tasks, checklist
- 5 SpecKit tools accessible through Copilot Chat
- Numbered feature specifications for all agents, tools, and skills
- Integration guide for systematic agent development
- specKitConstitution - Access constitutional framework
- specKitSpecTemplate - Feature specification template
- specKitPlanTemplate - Implementation planning template
- specKitTasksTemplate - Task breakdown template
- specKitChecklist - Quality validation checklist
- resourceDiscovery - List all available agents, prompts, skills, instructions, and templates
Total: 12 custom tools (6 core + 5 SpecKit + 1 discovery)
- spec-driven.md - Complete SDD guide
- agents.md - Agent integration checklist
- Constitutional Framework - Governance principles
- Templates - SDD templates with usage guide
- Specifications - Feature specs directory
- 6 VSCode native tools for all agents
- Code analysis, test generation, documentation, performance profiling, dependency analysis, and API design
- Production-ready error handling
- 3 new instruction sets: Go, Rust, Java
- Comprehensive best practices for each language
- Auto-applied based on file type
- Tool reference validation
- Improved error messages with suggestions
- Performance warnings for tool count
- Quality checks for names and descriptions
For detailed documentation on each agent, prompt, and skill, see the index.md file.
To package this extension yourself:
# Install vsce globally
npm install -g @vscode/vsce
# Package the extension
cd /path/to/agent-pro
vsce package
# This creates agent-pro-1.0.0.vsixContributions are welcome. Please see CONTRIBUTING.md for guidelines.
MIT License - see LICENSE file for details.
- Repository: github.com/mdaashir/agent-pro
- Issues: github.com/mdaashir/agent-pro/issues
- VS Code: 1.90.0 or higher
- GitHub Copilot: Active subscription required
- GitHub Copilot Chat: Extension must be installed
resources/agents-collection/- 24 expert agent definitionsresources/prompts-collection/- 5 reusable prompt templatesresources/instructions/- 5 automatic instruction setsresources/skills-collection/- 5 comprehensive skill bundles- Global access via Copilot Chat (
@agent-name) - Installed to
~/.config/Code/User/globalStorage/mdaashir.agent-pro/on activation