sf-agent-framework
Advanced tools
AI Agent Orchestration Framework for Salesforce Development - Two-phase architecture with 70% context reduction
AI-Powered Salesforce Development Framework | 70% Less Context | 3x Faster Delivery
🚀 Quick Start | 📚 Documentation | 💡 Features | 🤖 Agents | 📦 Installation
SF-Agent Framework is an intelligent orchestration system that transforms Salesforce development through AI agents. It dramatically reduces development time and improves code quality by leveraging specialized AI agents that understand Salesforce best practices.
| Benefit | Impact | How It Works |
|---|---|---|
| 70% Less Context | Faster AI responses, lower costs | Two-phase architecture separates planning from coding |
| 3x Faster Development | 20-minute story implementation | Lean agents optimized for specific tasks |
| 85% First-Pass Success | Less rework, higher quality | Agents trained on Salesforce best practices |
| 30+ Specialized Agents | Complete coverage | Each agent masters specific Salesforce domains |
| Enterprise-Ready | Production-safe | Built-in security, compliance, and governance |
Separate planning from coding for maximum efficiency:
PLANNING PHASE (128k context) DEVELOPMENT PHASE (32k context)
┌─────────────────────┐ ┌──────────────────────┐
│ • Full requirements │ │ • Focused stories │
│ • Architecture │ ────→ │ • Fast execution │
│ • Complex analysis │ │ • Lean agents │
│ • Story generation │ │ • 3x faster │
└─────────────────────┘ └──────────────────────┘
Transform requirements into actionable, self-contained stories:
# STORY-001: Implement Customer Portal Login
## Context
All information needed for implementation...
## Requirements
- User authentication via OAuth 2.0
- Session management with 30-min timeout
- Multi-factor authentication support
## Technical Specifications
- Apex controller: CustomerPortalController
- LWC component: customerLogin
- Integration: Auth0 SSO
Pre-built workflows for common scenarios:
# Global installation (recommended)
npm install -g sf-agent-framework@latest
# Verify installation
sfa --version
# Output: 4.0.1
# Or use directly with npx
npx sf-agent-framework install
# Create new project
sfa init my-salesforce-project
cd my-salesforce-project
# Or add to existing project
cd existing-project
sfa install
# Start interactive workflow
sfa workflow
# Or use agents directly
sfa phase planning
/sf-business-analyst *requirements "customer portal"
/sf-architect *design
# Switch to development
sfa phase development
/sf-developer-lean *implement
/sf-qa-lean *test
/sf-devops-lead-lean *deploy
| Command | Description | Example |
|---|---|---|
sfa init [name] | Create new project | sfa init my-app |
sfa workflow | Start interactive workflow | sfa workflow |
sfa phase [mode] | Switch context mode | sfa phase planning |
sfa story create | Generate stories | sfa story create --from requirements.md |
sfa help | Show all commands | sfa help |
sfa phase planning
AI agents analyze requirements with full context to create comprehensive plans.
sfa story create --from requirements.md
Requirements are broken into self-contained, implementable stories.
sfa phase development
Switch to optimized agents that work 3x faster with focused context.
/sf-developer-lean *implement
Lean agents implement each story with all necessary context included.
| Agent | Specialization | Commands |
|---|---|---|
sf-business-analyst | Requirements gathering | *requirements, *analysis |
sf-architect | Solution architecture | *design, *patterns |
sf-technical-architect | Technical design | *integration, *api |
sf-data-architect | Data modeling | *datamodel, *migration |
sf-security-architect | Security framework | *security, *compliance |
| Agent | Context Size | Focus | Speed |
|---|---|---|---|
sf-developer-lean | 32k | Apex, LWC, Triggers | 3x faster |
sf-admin-lean | 24k | Configuration, Fields | 2.5x faster |
sf-qa-lean | 24k | Testing, Quality | 2x faster |
sf-devops-lead-lean | 30k | Deployment, CI/CD | 2x faster |
Start with sfa workflow and choose:
$ sfa workflow
? Select your project type: New Salesforce Implementation
? Choose your approach: Requirements-driven development
? Select clouds: Sales Cloud, Service Cloud
✅ Workflow initialized: salesforce-greenfield-fullstack-v2
📋 Generated 12 implementation stories
🚀 Ready to start development!
AI-powered document generation that adapts to your needs:
| Template | Purpose | Time Saved |
|---|---|---|
user-story-enhanced | User stories with acceptance criteria | 20 min → 2 min |
technical-design-enhanced | Architecture documents | 60 min → 10 min |
test-class-enhanced | Complete test classes | 30 min → 5 min |
apex-class-enhanced | Apex classes with best practices | 15 min → 3 min |
lwc-component-enhanced | LWC with HTML, JS, metadata | 25 min → 5 min |
# Interactive mode
sfa template process user-story-enhanced
# With variables
sfa template process technical-design-enhanced \
--var project_name="Customer Portal" \
--var complexity="high"
your-salesforce-project/
├── .sf-core/ # Framework core (auto-managed)
│ ├── agents/ # 30+ specialized agents
│ ├── workflows/ # Enterprise workflows
│ └── templates/ # Document templates
├── src/ # Your Salesforce code
│ ├── classes/ # Apex classes
│ ├── lwc/ # Lightning Web Components
│ ├── objects/ # Custom objects
│ └── triggers/ # Apex triggers
├── docs/ # Generated documentation
│ ├── stories/ # Implementation stories
│ ├── architecture/ # Design documents
│ └── handoffs/ # Agent handoff packages
└── config/ # Configuration files
└── sf-agent.config.json
Break large requirements into manageable stories:
sfa shard docs/100-page-requirements.md
# Output: 20 focused implementation stories
Seamless collaboration between AI agents:
sfa handoff create --from sf-architect --to sf-developer
# Packages all context for smooth transition
Use with ChatGPT, Claude, or Gemini:
sfa build --teams-only
# Upload dist/teams/salesforce-core-team.txt to any AI chat
Works with all major IDEs:
✅ Greenfield Implementations - New Salesforce orgs ✅ Feature Development - Complex features for existing orgs ✅ Lightning Migration - Classic to Lightning Experience ✅ DevOps Setup - CI/CD pipelines and automation ✅ Package Development - Managed/unmanaged packages ✅ Integration Projects - Connect with external systems ✅ Data Migration - Move data between orgs
We welcome contributions! Areas where you can help:
See Contributing Guide for details.
MIT License - See LICENSE file for details.
| Feature | SF-Agent | Traditional Dev | Generic AI |
|---|---|---|---|
| Salesforce Expertise | ✅ Deep platform knowledge | ✅ Manual expertise | ❌ No context |
| Development Speed | ✅ 3x faster | ❌ Standard pace | ⚠️ Variable |
| Context Efficiency | ✅ 70% reduction | ❌ Full context | ❌ Massive usage |
| Code Quality | ✅ 85% first-pass | ⚠️ Varies | ⚠️ Heavy review |
| Best Practices | ✅ Built-in | ⚠️ Manual | ❌ Unaware |
| Cost | ✅ Lower AI costs | ❌ High dev costs | ❌ High tokens |
npm install -g sf-agent-framework@latest
Join thousands of developers building smarter with AI
FAQs
AI Agent Orchestration Framework for Salesforce Development - Two-phase architecture with 70% context reduction
The npm package sf-agent-framework receives a total of 10 weekly downloads. As such, sf-agent-framework popularity was classified as not popular.
We found that sf-agent-framework demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Security News
OpenSSF has issued a high-severity advisory warning open source developers of an active Slack-based campaign using impersonation to deliver malware.
Research
/Security News
Malicious packages published to npm, PyPI, Go Modules, crates.io, and Packagist impersonate developer tooling to fetch staged malware, steal credentials and wallets, and enable remote access.
Security News
Microsoft has released an open source toolkit for enforcing runtime security policies on AI agents as adoption accelerates faster than governance controls.