agentforce-adlc: Complete Setup Guide

Last Updated: October 01, 2026

🤖 AI Snapshot / TL;DR: The agentforce-adlc is a production-ready, open-source set of workflows designed to enhance AI coding assistants like Claude Code, Claude Desktop, and Cursor. It provides specialized context, rules, and prompts to improve code generation accuracy and consistency, backed by ⭐ 111 GitHub stars.

🔍 What is agentforce-adlc?

Agent development life cycle — build, deploy, test, and optimize agentforce agents using claude code skills and agent script dsl.. By providing structured, pre-defined instructions to your AI agent, it bridges the gap between generic AI responses and highly specialized, project-specific expertise. Instead of repeatedly explaining your tech stack or coding standards, you can simply drop these files into your project root.

✨ Key Features

• Direct authoring — the coding agent generates .agent files natively, not via markdown-to-agent conversion**
• Full lifecycle — author, discover, scaffold, deploy, test, and optimize in one toolchain**
• Safety built-in — llm-driven safety review across the entire lifecycle (authoring, deploy, test, optimize)**
• Deterministic agents — agent script dsl enforces code-level guarantees (conditionals, guards, transitions)**
• Session trace analysis — extract stdm data from data cloud for data-driven optimization**

🎯 Real-World Use Cases for agentforce-adlc

• End-to-End Feature Development: Chain multiple AI actions together (e.g., "Write tests → Implement feature → Update docs") in a single command.
• CI/CD Integration: Automate the generation of release notes, changelogs, or deployment scripts based on recent git commits.
• Complex Debugging: Provide the AI with a structured workflow to isolate, reproduce, and fix elusive bugs step-by-step.

⚙️ Installation & Configuration

Prerequisites

• An AI coding assistant that supports custom rules/skills (e.g., Claude Desktop, Claude Code, Cursor, Windsurf).
• A local project directory where you want to apply these rules.

Step 1: Download the Files

```bash

Clone the repository

git clone https://github.com/SalesforceAIResearch/agentforce-adlc.git cd agentforce-adlc ```

Step 2: Integrate into Your Project

Copy the relevant files to the root of your project:

• CLAUDE.md or claude.md → For Claude Desktop / Claude Code
• .cursorrules → For Cursor IDE
• .windsurfrules → For Windsurf (Codeium)

⚙️ Advanced Integration Options

For advanced setups, you can combine this workflows with other tools or customize its behavior:

```bash

Example: Linking .cursorrules globally (macOS/Linux)

ln -s /path/to/agentforce-adlc/.cursorrules ~/.cursorrules

Example: Using with Claude Code CLI

claude-code --rules /path/to/agentforce-adlc/claude.md ``` Note: Always review the rules or prompts before applying them to production codebases to ensure they align with your team's specific standards.

💡 Real-World Example Prompts

Once integrated, you can trigger these capabilities with simple prompts:

1. "Review this component using the project's established workflows."
2. "Generate a new API endpoint following the patterns defined in the workflows."
3. "Refactor this legacy code to align with the architectural rules provided."

⚖️ How agentforce-adlc Compares to Generic Alternatives

Feature	agentforce-adlc	Generic AI Prompts	Manual Coding
Context Awareness	✅ High (Project-specific)	❌ Low (Generic)	✅ High
Setup Time	< 2 minutes (Copy/Paste)	5-10 minutes per session	Hours
Consistency	✅ Enforced automatically	⚠️ Depends on user memory	✅ High
Best For	Team-wide AI standardization	One-off quick tasks	Highly complex logic

🛠️ Common Issues & Troubleshooting

• AI Ignoring the Rules: Ensure the file is named exactly .cursorrules or CLAUDE.md and is placed in the root of your project directory.
• Token Limit Exceeded: If the workflows file is very large, it might consume too many context tokens. Try splitting it into smaller, modular files.
• Conflicting Instructions: If you have multiple rule files, the AI might get confused. Consolidate them or use clear, hierarchical naming conventions.

🔄 Related AI Skills & Rules

If you found this useful, explore other complementary tools in our directory:

• Awesome MCP Servers (⭐ 95,665)
• GitHub MCP Server (⭐ 33,273)
• Explore Main Skills Directory

❓ Frequently Asked Questions (FAQ)

Q: Is agentforce-adlc free to use?
A: Yes, this is an open-source AI skill/rule set. You can freely copy and adapt it for personal or commercial projects.

Q: Which AI tools support this workflows?
A: Depending on the files included, it supports Claude Desktop, Claude Code, Cursor, Windsurf, and other rule-based AI editors.

Q: How do I update this workflows?
A: Simply pull the latest changes from the GitHub repository and replace the files in your project's root directory.