SalesforceAIResearch

SalesforceAIResearch / agentforce-adlc

Public

Agent Development Life Cycle — Build, deploy, test, and optimize Agentforce agents using Claude Code skills and Agent Script DSL.

agentforce claude-skill cursor-rules
13
1
100% credibility
Found Apr 17, 2026 at 13 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A set of AI coding skills for building, testing, deploying, and improving intelligent agents in Salesforce Agentforce.

How It Works

1
🔍 Discover the toolkit

You hear about a helpful set of tools from Salesforce that lets AI build smart customer service agents for you.

2
📥 Add to your coding helper

You easily add these special instructions to your AI coding friend like Claude or Cursor with a simple download.

3
💭 Describe your dream agent

You chat with your AI friend and explain what your agent should do, like helping customers check orders or process returns.

4
✨ AI creates everything

Your AI friend writes the agent's instructions, checks your setup, creates any missing pieces, and makes sure it's safe.

5
🧪 Test it out

You run quick tests with sample questions to see if the agent understands and responds correctly.

6
🚀 Launch to your team

With one command, you send the agent live in your Salesforce world where it starts helping real people.

🎉 Agent is live and smart

Your new agent handles customer chats smoothly, learns from real use, and you tweak it anytime to make it even better.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 13 to 13 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is agentforce-adlc?

agentforce-adlc is an agent development kit for the full Agent Development Life Cycle (ADLC) of Salesforce Agentforce agents. It equips Claude Code with skills to author .agent files in Agent Script DSL, discover org resources, scaffold Flows and Apex stubs, deploy via sf CLI, run preview/batch tests, and optimize from session traces in Data Cloud. Python-based, it installs as Claude Code plugins or Cursor skills, with standalone scripts for discovery and scaffolding.

Why is it gaining traction?

Unlike generic agent github claude tools or agent github copilot vscode extensions, it handles Agentforce end-to-end: direct .agent generation without markdown conversion, LLM-driven safety reviews across phases, and deterministic logic via DSL for reliable transitions. Developers notice the integrated pipeline—safety gates block deploys, hooks validate syntax on writes, and trace analysis spots production issues fast. It's a niche agent development environment tailored for Salesforce, saving hours on boilerplate.

Who should use this?

Salesforce platform engineers building service agents for order status, returns, or escalations. Teams using Claude Code or Cursor for agent github code who need structured workflows with org discovery, auto-stubs for missing actions, and guardrails against prompt injections or unsafe behaviors. Ideal for devs evaluating agent development kit rag or agent development kit mcp patterns in production orgs.

Verdict

Grab it if you're deep in Agentforce—solid for rapid prototyping despite 13 stars and 1.0% credibility score signaling early maturity. Docs are thorough, tests cover 88 cases, but expect tweaks for edge deploys. Worth the install for Claude-powered ADLC.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.