Salesforce-focused compound engineering plugin for Claude Code, Cursor, and Codex. Skills-first V3 architecture with parallel agent dispatch, institutional knowledge compounding, and Apex/LWC/Flow/Integration coverage.
Salesforce Hosted MCP Server setup, configuration, URL patterns, ECA management, security model, complete tool reference, and troubleshooting
Salesforce integration patterns for REST, SOAP, Platform Events, and external system connectivity
Set up the sf-compound-engineering-plugin environment for a new Salesforce developer. Validates Salesforce CLI, sfdx-project.json, MCP servers, and Claude Code plugin install. Use when onboarding to the plugin or troubleshooting install. Trigger phrases: 'set up the plugin', 'check plugin prerequisites', 'verify my Salesforce environment'.
Extract conversation skeleton and error signals from a single Claude/Codex/Cursor session file. Internal helper for sf-sessions; not usually invoked directly.
Discover Claude/Codex/Cursor session files for the current Salesforce repo and extract metadata. Internal helper for sf-sessions; not usually invoked directly. Trigger phrase: 'list sessions for this repo'.
External network access
Connects to servers outside your machine
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Instruction-Based Compound Engineering for Salesforce — a multi-platform plugin (Claude Code, Cursor, Codex, and 9 other AI coding tools) where each iteration becomes smarter than the last through institutional knowledge capture and parallel persona dispatch.
V3 is a skills-first architecture. Commands were retired — every entry point is now a skill that auto-routes from natural-language phrases via its
descriptionfrontmatter, with direct slash invocation (/sf-<name>) still supported. SeeCHANGELOG.mdfor the migration guide from v2.x.
┌─ HUMAN (taste) ─┐ ┌──────── AI (in the loop) ────────┐ ┌─ HUMAN (taste) ─┐
Ideate · Brainstorm → Plan(40%) · Deepen · Work(20%) · Review · Resolve → Polish → Compound → Repeat
• Ideate — decide what's worth building (/sf-ideate)
• Brainstorm — explore requirements through collaborative dialogue (/sf-brainstorm)
• Plan — research & design using 60 skills + parallel research personas (/sf-plan)
• Deepen — enhance the plan with section-level parallel research (/sf-deepen)
• Work — implement with pre-research + system-wide test checks (/sf-work)
• Review — parallel persona dispatch across 60 specialist personas (/sf-review)
• Polish — taste pass via /sf-polish: SLDS2/UX, accessibility (WCAG), copy (UI surfaces only)
• Compound — capture learnings to docs/solutions/, personas, skills, CLAUDE.md (/sf-compound)
The sandwich. Humans own the two ends — Ideate (what's worth building) and Polish (does it feel right) — the "bread". The AI runs the middle "filling" in the loop. As models get better at execution, human attention concentrates where machines are still weak: taste and judgment.
All nine core entry points (
/sf-ideate,/sf-brainstorm,/sf-plan,/sf-deepen,/sf-work,/sf-review,/sf-polish,/sf-compound,/sf-lfg) are skills in V3 — they auto-route from natural-language phrases via theirdescriptionfrontmatter, and direct slash invocation continues to work.
Above the loop, /sf-strategy maintains an optional repo-root STRATEGY.md (target problem, approach, users, key metrics, tracks) that sf-ideate, sf-brainstorm, and sf-plan read as grounding when it exists.
Each iteration starts smarter because learnings compound into docs/solutions/, personas, skills, and CLAUDE.md.
Principles. This plugin is opinionated. Seven principles — preserve the quality ceiling, verifiability, stay in the loop, the spec is the artifact, taste over typing, agent-native docs, outsource thinking not understanding — govern every skill and every review. See
PRINCIPLES.md. Each core workflow skill declares which principles it enforces.
# Add as a Claude Code plugin marketplace
/plugin marketplace add https://github.com/gellasangameshgupta/salesforce-compound-engineering-plugin
# Install
/plugin install sf-compound-engineering
# GitHub Copilot
bunx @gellasangameshgupta/sf-compound-plugin install sf-compound-engineering --to copilot
# Cursor
bunx @gellasangameshgupta/sf-compound-plugin install sf-compound-engineering --to cursor
# Windsurf
bunx @gellasangameshgupta/sf-compound-plugin install sf-compound-engineering --to windsurf
# Gemini CLI
bunx @gellasangameshgupta/sf-compound-plugin install sf-compound-engineering --to gemini
# OpenCode
bunx @gellasangameshgupta/sf-compound-plugin install sf-compound-engineering --to opencode
# Codex
bunx @gellasangameshgupta/sf-compound-plugin install sf-compound-engineering --to codex
# Kiro
bunx @gellasangameshgupta/sf-compound-plugin install sf-compound-engineering --to kiro
# Factory Droid
bunx @gellasangameshgupta/sf-compound-plugin install sf-compound-engineering --to droid
# Pi
bunx @gellasangameshgupta/sf-compound-plugin install sf-compound-engineering --to pi
# OpenClaw
bunx @gellasangameshgupta/sf-compound-plugin install sf-compound-engineering --to openclaw
# Qwen Code
bunx @gellasangameshgupta/sf-compound-plugin install sf-compound-engineering --to qwen
# Auto-detect and install to all detected tools
bunx @gellasangameshgupta/sf-compound-plugin install sf-compound-engineering --to all
# Sync from current directory to all detected tools
bunx @gellasangameshgupta/sf-compound-plugin sync
npx claudepluginhub gellasangameshgupta/salesforce-compound-engineering-plugin --plugin sf-compound-engineeringComprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
Harness-native ECC plugin for engineering teams - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, MCP conventions, and operator workflows for Claude Code plus adjacent agent harnesses
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Browser automation and end-to-end testing MCP server by Microsoft. Enables Claude to interact with web pages, take screenshots, fill forms, click elements, and perform automated browser testing workflows.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.