Compounding Engineering Plugin

June 15, 2026 · View on GitHub

AI-powered development tools that get smarter with every use. Make each unit of engineering work easier than the last.

Getting Started

After installing, run /ce-setup in any project. It diagnoses your environment, installs missing tools, and bootstraps project config in one interactive flow.

Components

ComponentCount
Agents50+
Skills38+

Skills

The primary entry points for engineering work, invoked as slash commands. Detailed user-facing documentation for many skills lives in docs/skills/ — each linked skill name below points to its page (purpose, novel mechanics, use cases, chain position). Skills without dedicated docs are still listed; their SKILL.md in the source tree is authoritative.

Core Workflow

ce-strategy anchors the loop upstream; ce-product-pulse closes it with a read on user outcomes.

SkillDescription
/ce-strategyCreate or maintain STRATEGY.md — the product's target problem, approach, persona, key metrics, and tracks. Re-runnable to update. Read as grounding by /ce-ideate, /ce-brainstorm, and /ce-plan when present
/ce-ideateOptional big-picture ideation: generate and critically evaluate grounded ideas, then route the strongest one into brainstorming. Writes the ranked ideation artifact as a single self-contained HTML file by default (human-facing); pass output:md for markdown (exclusive — html OR md, never both)
/ce-brainstormInteractive Q&A to think through a feature or problem and write a right-sized requirements doc before planning. Pass output:html to write the doc as a single self-contained HTML file instead of markdown (exclusive — md OR html, never both)
/ce-planCreate structured plans for any multi-step task -- software features, research workflows, events, study plans -- with automatic confidence checking. Pass output:html to write the plan as a single self-contained HTML file instead of markdown (exclusive — md OR html, never both)
/ce-code-reviewStructured code review with tiered persona agents, confidence gating, and dedup pipeline
/ce-workExecute work items systematically
/ce-debugSystematically find root causes and fix bugs -- traces causal chains, forms testable hypotheses, and implements test-first fixes
/ce-compoundDocument solved problems to compound team knowledge
/ce-compound-refreshRefresh stale or drifting learnings and decide whether to keep, update, replace, or archive them
/ce-optimizeRun iterative optimization loops with parallel experiments, measurement gates, and LLM-as-judge quality scoring
/ce-product-pulseGenerate a single-page, time-windowed report on usage, performance, errors, and followups. Saves reports to docs/pulse-reports/ as a browseable timeline of what users experienced

Research & Context

SkillDescription
/ce-sessionsAsk questions about session history across Claude Code, Codex, and Cursor
/ce-slack-researchSearch Slack for interpreted organizational context -- decisions, constraints, and discussion arcs
ce-riffrec-feedback-analysisConvert Riffrec recordings, videos, audio, or notes into structured feedback. Routes between setup, quick bug report, and extensive analysis that hands off to ce-brainstorm

Git Workflow

SkillDescription
ce-clean-gone-branchesClean up local branches whose remote tracking branch is gone
ce-commitCreate a git commit with a value-communicating message
ce-commit-push-prCommit, push, and open a PR with an adaptive description; also update an existing PR description, or generate a description on its own without committing
ce-worktreeEnsure work happens in an isolated git worktree — detect existing isolation, prefer native worktree tooling, else create one

Workflow Utilities

SkillDescription
/ce-demo-reelCapture a visual demo reel (GIF demos, terminal recordings, screenshots) for PRs with project-type-aware tier selection
/ce-promoteDraft user-facing announcement copy for a shipped feature (X post, changelog blurb, LinkedIn, email); voice-matched via the Spiral CLI when installed, a lite layer of editorial & social expertise without it
/ce-report-bugReport a bug in the compound-engineering plugin
/ce-resolve-pr-feedbackResolve PR review feedback in parallel
/ce-test-browserRun browser tests on PR-affected pages
/ce-test-xcodeBuild and test iOS apps on simulator using XcodeBuildMCP
/ce-setupDiagnose environment, install missing tools, and bootstrap project config
/ce-updateCheck compound-engineering plugin version and fix stale cache (Claude Code only)
/ce-release-notesSummarize recent compound-engineering plugin releases, or answer a question about a past release with a version citation

Development Frameworks

SkillDescription
ce-agent-native-architectureBuild AI agents using prompt-native architecture
ce-dhh-rails-styleWrite Ruby/Rails code in DHH's 37signals style
ce-frontend-designCreate production-grade frontend interfaces
ce-polishConversational UX polish — start a dev server, open the feature in a browser, and iterate together; auto-detects 8 frameworks. Manual invocation only

Review & Quality

SkillDescription
ce-doc-reviewReview documents using parallel persona agents for role-specific feedback
/ce-simplify-codeSimplify recent code changes for reuse, quality, and efficiency — parallel reviewers find issues, fixes applied, behavior verified by tests

Content & Collaboration

SkillDescription
ce-proofCreate, edit, and share documents via Proof collaborative editor

Automation & Tools

SkillDescription
ce-gemini-imagegenGenerate and edit images using Google's Gemini API

Beta / Experimental

SkillDescription
ce-dogfood-betaDiff-scoped browser QA of the active branch: builds an exhaustive test matrix of every change, drives the app with agent-browser, then auto-fixes issues, adds regression tests, and commits each fix until green
/lfgFull autonomous engineering workflow

Agents

Agents are specialized subagents invoked by skills — you typically don't call these directly.

Review

AgentDescription
ce-agent-native-reviewerVerify features are agent-native (action + context parity)
ce-api-contract-reviewerDetect breaking API contract changes
ce-architecture-strategistAnalyze architectural decisions and compliance
ce-code-simplicity-reviewerFinal pass for simplicity and minimalism
ce-correctness-reviewerLogic errors, edge cases, state bugs
ce-data-integrity-guardianDatabase migrations and data integrity
ce-data-migration-reviewerSchema drift, migration safety, mapping verification, deploy-window checks
ce-deployment-verification-agentCreate Go/No-Go deployment checklists for risky data changes
ce-julik-frontend-races-reviewerReview JavaScript/Stimulus code for race conditions
ce-maintainability-reviewerCoupling, complexity, naming, dead code
ce-pattern-recognition-specialistAnalyze code for patterns and anti-patterns
ce-performance-oraclePerformance analysis and optimization
ce-performance-reviewerRuntime performance with confidence calibration
ce-reliability-reviewerProduction reliability and failure modes
ce-security-reviewerExploitable vulnerabilities with confidence calibration
ce-security-sentinelSecurity audits and vulnerability assessments
ce-swift-ios-reviewerSwift and iOS code review -- SwiftUI state, retain cycles, concurrency, Core Data threading, accessibility
ce-testing-reviewerTest coverage gaps, weak assertions
ce-project-standards-reviewerCLAUDE.md and AGENTS.md compliance
ce-adversarial-reviewerConstruct failure scenarios to break implementations across component boundaries

Document Review

AgentDescription
ce-coherence-reviewerReview documents for internal consistency, contradictions, and terminology drift
ce-design-lens-reviewerReview plans for missing design decisions, interaction states, and AI slop risk
ce-feasibility-reviewerEvaluate whether proposed technical approaches will survive contact with reality
ce-product-lens-reviewerChallenge problem framing, evaluate scope decisions, surface goal misalignment
ce-scope-guardian-reviewerChallenge unjustified complexity, scope creep, and premature abstractions
ce-security-lens-reviewerEvaluate plans for security gaps at the plan level (auth, data, APIs)
ce-adversarial-document-reviewerChallenge premises, surface unstated assumptions, and stress-test decisions

Research

AgentDescription
ce-best-practices-researcherGather external best practices and examples
ce-framework-docs-researcherResearch framework documentation and best practices
ce-git-history-analyzerAnalyze git history and code evolution
ce-issue-intelligence-analystAnalyze GitHub issues to surface recurring themes and pain patterns
ce-learnings-researcherSearch institutional learnings for relevant past solutions
ce-repo-research-analystResearch repository structure and conventions
ce-session-historianSearch prior Claude Code, Codex, and Cursor sessions for related investigation context
ce-slack-researcherSearch Slack for organizational context relevant to the current task
ce-web-researcherPerform iterative web research and return structured external grounding (prior art, adjacent solutions, market signals, cross-domain analogies)

Design

AgentDescription
ce-design-implementation-reviewerVerify UI implementations match Figma designs
ce-design-iteratorIteratively refine UI through systematic design iterations
ce-figma-design-syncSynchronize web implementations with Figma designs

Workflow

AgentDescription
ce-pr-comment-resolverAddress PR comments and implement fixes
ce-spec-flow-analyzerAnalyze user flows and identify gaps in specifications

Docs

AgentDescription
ce-ankane-readme-writerCreate READMEs following Ankane-style template for Ruby gems

Installation

See the repo root Install section for current installation instructions across Claude Code, Codex, Cursor, Copilot, Droid, Qwen, and converter-backed targets.

Then run /ce-setup to check your environment and install recommended tools.

Version History

See the repo root CHANGELOG.md for canonical release history.

License

MIT