agents
Health Warn
- License — License: MIT
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Low visibility — Only 7 GitHub stars
Code Pass
- Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Pass
- Permissions — No dangerous permissions requested
This project is an extensive plugin ecosystem for Claude Code that turns it into a supervised, multi-agent development environment. It uses specialized agents, scripts, and real-time hooks to enforce quality standards, prevent code duplication, and inject framework-specific knowledge into your workflow.
Security Assessment
The automated code scan reviewed 12 files and found no dangerous patterns, hardcoded secrets, or requests for dangerous permissions. However, the tool inherently works by intercepting Write, Edit, and Bash calls to enforce rules, which requires local system access. Additionally, the quick install relies on executing a shell setup script (`setup.sh`). Because the project has low visibility, manual review of this setup script and the 82 hooks is recommended to ensure they align with your system security policies before running them. Overall risk is rated as Medium due to the highly invasive nature of the hooks and shell execution required for setup, despite the clean automated scan.
Quality Assessment
The project is actively maintained, with its last push occurring today. It is released under the standard and permissive MIT license. However, community trust and visibility are currently very low; it only has 7 GitHub stars, meaning the codebase has not been broadly vetted by a large audience.
Verdict
Use with caution — the automated scan is clean, but the low community engagement and the execution of local hooks and shell scripts warrant a quick manual review before integrating into your workflow.
Redefining development through cognitive automation and collaborative agent systems.
Fusengine Claude Code Plugins
A plugin ecosystem that turns Claude Code into a supervised, multi-agent development environment. Expert agents write code, hooks enforce quality in real-time, skills inject framework-specific knowledge, and intelligent cartography auto-maps plugins and projects — so Claude never guesses, never duplicates, and always follows your architecture.

What It Does
Without plugins: Claude Code writes code based on its training data. It can hallucinate APIs, duplicate existing code, ignore your project structure, and produce inconsistent quality.
With Fusengine plugins:
- Expert agents detect your project type (Next.js, Laravel, React, Astro, Swift...) and load framework-specific documentation via MCP servers before writing a single line
- 82 hooks intercept every Write/Edit/Bash call in real-time to enforce file size limits (<100 lines), block code duplication (DRY), require SOLID references, and validate security
- 125 skills inject copy-paste-ready templates, architecture patterns, and best practices directly into agent context — no hallucination needed
- APEX workflow structures every task through Analyze → Plan → Execute (TDD) → Review → Validate — preventing the "just write code and hope" approach
- Sniper validation runs a 7-phase quality check after every modification: explore → research → grep usages → lint → fix → zero errors
Quick Install
# Add marketplace
/plugin marketplace add fusengine/agents
# Install all plugins
/plugin install fuse-ai-pilot fuse-commit-pro fuse-laravel fuse-nextjs fuse-react fuse-astro fuse-swift-apple-expert fuse-solid fuse-tailwindcss fuse-design fuse-prompt-engineer fuse-shadcn-ui fuse-security fuse-changelog
# Setup (hooks + API keys + MCP servers)
~/.claude/plugins/marketplaces/fusengine-plugins/setup.sh # macOS / Linux
~\.claude\plugins\marketplaces\fusengine-plugins\setup.ps1 # Windows
Statusline (optional):
bun --cwd ~/.claude/plugins/marketplaces/fusengine-plugins/plugins/core-guards/statusline run config
How It Works
User prompt → Hook detects project type → Expert agent activated
→ Hook loads SOLID references → Agent reads docs via MCP
→ Hook blocks if DRY violation → Agent writes code
→ Hook checks file size → Sniper validates quality
→ Hook blocks secrets → Commit with version bump
Every step is intercepted. Claude cannot skip research, cannot duplicate code, cannot exceed file size limits, and cannot commit without security validation.
Plugins
Development — Framework Expert Agents
Each plugin provides an expert agent that auto-activates when it detects the framework in your project. The agent has access to official documentation via MCP servers and follows SOLID principles enforced by hooks.
| Plugin | Detects | What the agent does |
|---|---|---|
| fuse-nextjs | next.config.* |
App Router, RSC, Prisma 7, Better Auth, proxy.ts patterns |
| fuse-laravel | composer.json + artisan |
Eloquent, Livewire, Blade, queues, Sanctum, Stripe Connect |
| fuse-react | package.json + React |
React 19 hooks, TanStack Router/Form, Zustand stores |
| fuse-astro | astro.config.* |
Islands, Content Layer, Actions, SEO, Starlight, i18n |
| fuse-swift-apple-expert | Package.swift |
SwiftUI, concurrency, all Apple platforms (iOS → visionOS) |
| fuse-tailwindcss | tailwind.config.* |
v4.1 CSS-first config, @theme, @utility, OKLCH colors |
| fuse-design | Any UI task | Gemini Design MCP + shadcn/ui + WCAG 2.2 accessibility |
| fuse-shadcn-ui | components.json |
Radix/Base UI detection, registry, theming, migration |
Quality & Security — Automated Enforcement
| Plugin | What it enforces |
|---|---|
| fuse-ai-pilot | APEX workflow orchestration, 7-phase sniper validation, DRY detection via jscpd, 4-level cache (60-75% token savings) |
| fuse-security | OWASP Top 10 scanning, CVE research via NVD/OSV.dev, dependency audit, secrets detection, auth patterns audit |
| fuse-solid | Files < 100 lines, interface separation, JSDoc mandatory — auto-detected for TypeScript, PHP, Swift, Go, Java, Ruby, Rust |
Productivity — Automation
| Plugin | What it automates |
|---|---|
| fuse-commit-pro | Conventional commits with security check, auto version bump, CHANGELOG, git tag — blocks secrets from commits |
| fuse-prompt-engineer | Prompt creation with CoT/Few-Shot/Meta-Prompting, A/B testing, 50+ template library, agent design |
| fuse-cartographer | Intelligent cartography: auto-generates navigable maps of plugins and projects at SessionStart with merge-preserving enrichment — agents navigate via linked index trees, /map --enrich completes descriptions from source frontmatter |
| fuse-changelog | Monitors Claude Code updates, detects breaking changes in plugins, gathers community feedback via Exa |
Core (auto-installed, always active)
| Plugin | What it guards |
|---|---|
| core-guards | Blocks git push --force, rm -rf, npm install without lock, enforces SOLID file limits, tracks session state, DRY duplication blocking |
| claude-rules | Injects APEX/SOLID/DRY rules into every prompt so Claude never forgets the methodology |
Key Features
| Feature | How it prevents mistakes |
|---|---|
| APEX Workflow | Forces Analyze before coding — no more "let me just write this real quick" |
| Agent Teams | Parallel agents with exclusive file ownership — no merge conflicts |
| DRY Detection | Blocks Write/Edit if function/class name already exists — forces import or shared extraction |
| SOLID Hooks | Denies file save if >100 lines or missing SOLID reference read |
| Sniper Validation | 7-phase post-edit check: explore → research → grep → lint → fix → zero errors |
| 4-Level Cache | Caches exploration, docs, lessons, tests — 60-75% token savings |
| 28 MCP Servers | Context7, Exa, Astro docs, Gemini Design, shadcn, Playwright, and more |
| Smart Commits | Security scan before commit, auto version bump, CHANGELOG, shields.io badge sync |
Documentation
| Section | Content |
|---|---|
| Getting Started | Installation, configuration, first steps |
| Workflow | APEX methodology, agents, skills, commands |
| Plugins | Per-plugin documentation and skills |
| Reference | Architecture, hooks, MCP servers, cache system |
License
MIT
Reviews (0)
Sign in to leave a review.
Leave a reviewNo results found