Compound Engineering Plugin

Project Overview

The Compound Engineering Plugin is a Claude Code plugin from Every that implements a systematic workflow for AI-assisted development.^[1] The core philosophy: each unit of engineering work should make subsequent units easier — not harder.

Created by Dan Shipper and Kieran Klaassen at Every, the plugin packages the exact workflow they use to run five products (Cora, Monologue, Sparkle, Spiral, and Every.to) with primarily single-person engineering teams.^[2]

As of June 2026, the repo has 21,000+ GitHub stars and 1,500+ forks — up from roughly 9,300 stars in February — making it one of the most popular Claude Code plugins for teams wanting structured AI workflows rather than ad-hoc prompting.^[1] Development is unusually active: v3.12.0 shipped June 9, 2026, the third release in the first week of June alone.^[3]

The Philosophy

Traditional development accumulates technical debt. Each feature adds complexity. After years, teams spend more time fighting their system than building on it.^[2]

Compound engineering inverts this:

• Features teach the system new capabilities instead of adding complexity
• Bug fixes eliminate entire categories of future bugs when patterns are captured
• Knowledge compounds so the codebase becomes easier to work with over time

The key ratio: 80% planning and review, 20% execution. Most thinking happens before and after code gets written — the opposite of how most developers work.

The Main Loop

Ideate → Brainstorm → Plan → Work → Review → Polish → Compound → Repeat

The loop has expanded since early 2026 from four steps to seven, with optional ideation and brainstorming up front and a polish pass before knowledge capture.^[2] The middle steps are familiar. The final step — Compound — is where gains accumulate. Skip it, and you've done traditional engineering with AI assistance. Commands have also been renamed: the old /workflows:* namespace is now /ce-* (e.g., /ce-plan, /ce-work, /ce-code-review, /ce-compound), plus /lfg for a fully autonomous idea-to-PR run.^[1]

1. Plan

Transform an idea into a blueprint:

• Understand the requirement (what, why, constraints)
• Research the codebase (existing patterns, similar functionality)
• Research externally (framework docs, best practices)
• Design the solution (approach, affected files)
• Validate the plan (completeness check)

The /ce-plan command spawns parallel research agents — repo analysis, framework docs, and best practices — then merges results into a structured plan with automatic confidence checking.^[1]

2. Work

Execution follows the plan:

• Set up isolation (git worktrees keep work separate)
• Execute the plan step by step
• Run validations after each change
• Track progress and handle issues

The /ce-work command creates worktrees, implements tasks, and optionally spawns quality-check agents.

3. Review

Multi-agent code review catches issues and captures learnings:

• Tiered persona agents review in parallel, with confidence gating and a dedup pipeline^[1]
• Findings prioritized as P1 (must fix), P2 (should fix), P3 (nice to fix)
• Agent fixes issues based on feedback
• Patterns captured for future prevention

The /ce-code-review command is the standout feature — specialized reviewers covering security, correctness, architecture, data integrity, and more running simultaneously.

4. Compound (The Key Step)

This is what separates compound engineering from traditional AI-assisted dev:

• Capture the solution — What worked? What didn't? What's reusable?
• Make it findable — Tag with metadata for future retrieval
• Update the system — Add patterns to CLAUDE.md, create new agents
• Verify the learning — Would the system catch this automatically next time?

What's in the Box

The plugin has grown substantially since February 2026 (then 26 agents, 23 commands, 13 skills). As of June 2026 it ships with:^[1]

50+ Specialized Agents:

• Code review specialists (security, correctness, architecture, data integrity, API contracts)
• Research agents (codebase, sessions, Slack)
• Design and frontend agents
• Workflow automation agents

38+ Skills (invoked as slash commands):

• /ce-strategy — Anchor work to a STRATEGY.md
• /ce-ideate and /ce-brainstorm — Clarify fuzzy requirements
• /ce-plan — Generate implementation blueprints
• /ce-work — Execute with progress tracking
• /ce-code-review — Tiered multi-agent code review
• /ce-compound and /ce-compound-refresh — Capture and maintain learnings
• /lfg — Fully autonomous idea-to-PR workflow
• Plus git workflow, debugging, browser testing, demo capture, and release-notes utilities

Review Agents Deep Dive

The /ce-code-review command spawns specialized ce- prefixed reviewers in parallel, with confidence gating and deduplication of findings:^[1]

Setup auto-detects your tech stack and configures the appropriate stack-specific review agents.

Installation

Claude Code (native):

/plugin marketplace add https://github.com/EveryInc/compound-engineering-plugin
/plugin install compound-engineering

After installing, run /ce-setup — it diagnoses your environment, installs missing tools, and bootstraps project config in one interactive flow.^[1]

Other platforms: Cursor and Codex have first-class install paths, and the bundled CLI (released in lockstep with the plugin) installs to GitHub Copilot, Factory Droid, Qwen Code, OpenCode, Pi, Gemini, and Kiro.^[2]

File Structure

After installation, projects get:^[2]

your-project/
├── CLAUDE.md           # Agent instructions, preferences, patterns
├── docs/
│   ├── brainstorms/    # /workflows:brainstorm output
│   ├── solutions/      # /workflows:compound output
│   └── plans/          # /workflows:plan output
└── todos/              # Review findings with priority/status
    ├── 001-ready-p1-fix-auth.md
    └── 002-pending-p2-add-tests.md

CLAUDE.md is the most important file — the agent reads it every session. Put preferences, patterns, and project context here. When something goes wrong, add a note so the agent learns.

docs/solutions/ builds institutional knowledge. Each solved problem becomes searchable documentation.

Strengths

• Systematic workflow — Structure beats ad-hoc prompting for complex projects
• Multi-agent review — Parallel specialists with confidence gating catch more than one generalist
• Knowledge capture — The Compound step prevents repeated mistakes
• Cross-platform — Works across major AI coding tools
• Battle-tested — Used internally at Every to run five products
• Open source — MIT licensed, full visibility into agents and prompts

Weaknesses / Considerations

• Overhead for small tasks — The full seven-step loop is overkill for quick fixes
• Claude Code first — Cursor and Codex are supported, but other platforms rely on the CLI installer
• Learning curve — The 80/20 ratio feels slow until it clicks, and the surface area (50+ agents, 38+ skills) keeps growing
• Token usage — Parallel review agents consume significant context
• Churn — A near-daily release cadence and a full command rename (/workflows:* → /ce-*) since February mean docs and habits go stale fast^[3]
• Opinionated — Assumes you want Every's workflow, not your own

What Developers Say

Independent commentary is still thin relative to the project's star count — most coverage comes from Every itself or interview features. One representative third-party endorsement:

"Compound engineering plugin for Claude Code developed by @kieranklaassen and @danshipper is basically my go-to plan mode in CC now, I rarely use the regular plan mode these days." — Soumitra Shukla on X^[4]

The methodology has also drawn mainstream practitioner attention: Peter Yang's Creator Economy interview with Klaassen, "How to Make Claude Code Better Every Time You Use It" (February 2026), walks through the full system as a tutorial.^[5]

Who Built It

Every is Dan Shipper's company focused on AI-augmented productivity. They run:

• Cora — AI computer assistant
• Monologue — Voice journaling
• Sparkle — AI writing tool
• Spiral — Writing assistant
• Every.to — Newsletter/publication

The compound engineering system emerged from running all five products with small teams. Kieran Klaassen is the primary engineer behind the plugin implementation.

Ideal User

• Solo developers or small teams managing complex codebases
• Claude Code power users wanting more structure
• Teams tired of repeated mistakes who want knowledge capture
• Developers who value planning over "just start coding"
• Anyone experimenting with multi-agent workflows

Bottom Line

The Compound Engineering Plugin represents a mature philosophy for AI-assisted development. The core insight — that each unit of work should make future work easier — is simple but rarely implemented systematically.

The 80/20 ratio (planning/review vs. execution) feels counterintuitive to developers trained to "just start coding," but Every's results running five products with small teams suggest it works.

The parallel persona review is the technical standout. Most AI coding tools review with one model in one pass. Compound Engineering throws specialized agents at security, correctness, architecture, and style simultaneously — with confidence gating and dedup to keep the output usable — catching more issues by construction.

Recommended for solo developers and small teams ready to invest in systematic AI workflows rather than ad-hoc prompting — this remains the most complete open-source option available, and it's free (MIT licensed). Not recommended if you want a lightweight assist for quick fixes or resist adopting someone else's workflow wholesale.

Outlook: Momentum is strong — stars more than doubled (9,300 → 21,000+) between February and June 2026, releases ship near-daily, and platform coverage now includes Cursor.^[1] The main risk is churn: the v3 restructure renamed every command, and the rapid pace means today's docs may not match next month's plugin. Expect continued expansion as Every dogfoods it across five products.

---