https://github.com/vibeeval/vibecosystem
Your AI software team. 137 agents, 269 skills, 53 hooks. Self-learning, multi-agent swarm, cross-project training. Built on Claude Code.
https://github.com/vibeeval/vibecosystem
ai-agents ai-software-team anthropic automation claude claude-code claude-skills developer-tools devtools hooks multi-agent open-source self-learning vibe-coding vibecoding
Last synced: 3 months ago
JSON representation
Your AI software team. 137 agents, 269 skills, 53 hooks. Self-learning, multi-agent swarm, cross-project training. Built on Claude Code.
- Host: GitHub
- URL: https://github.com/vibeeval/vibecosystem
- Owner: vibeeval
- License: mit
- Created: 2026-03-14T11:57:55.000Z (4 months ago)
- Default Branch: main
- Last Pushed: 2026-04-02T09:55:05.000Z (3 months ago)
- Last Synced: 2026-04-02T23:58:13.136Z (3 months ago)
- Topics: ai-agents, ai-software-team, anthropic, automation, claude, claude-code, claude-skills, developer-tools, devtools, hooks, multi-agent, open-source, self-learning, vibe-coding, vibecoding
- Language: C#
- Size: 7.2 MB
- Stars: 373
- Watchers: 8
- Forks: 29
- Open Issues: 5
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Security: SECURITY.md
- Agents: AGENTS.md
Awesome Lists containing this project
- awesome-claude-code - **vibecosystem** - (507 β) - An AI software-team system for Claude Code with agents, skills, hooks, and self-learning workflow support. (π§ Agent Skills)
README
# vibecosystem
**Your AI software team. Built on Claude Code.**
[](https://opensource.org/licenses/MIT)
[](#agents)
[](#skills)
[](#hooks)
[](#rules)
[](https://github.com/vibeeval/vibecosystem/actions/workflows/validate.yml)
[](#multi-cli)
[](#multi-cli)
[](#multi-cli)
[Turkce](#turkce) | [English](#english) | [Espanol](docs/README_ES.md) | [Francais](docs/README_FR.md) | [Deutsch](docs/README_DE.md) | [Portugues](docs/README_PT.md) | [Italiano](docs/README_IT.md) | [Nederlands](docs/README_NL.md) | [δΈζ](docs/README_ZH.md) | [ζ₯ζ¬θͺ](docs/README_JA.md) | [νκ΅μ΄](docs/README_KO.md) | [Ψ§ΩΨΉΨ±Ψ¨ΩΨ©](docs/README_AR.md) | [ΰ€Ήΰ€Ώΰ€¨ΰ₯ΰ€¦ΰ₯](docs/README_HI.md) | [Π ΡΡΡΠΊΠΈΠΉ](docs/README_RU.md)
vibecosystem turns Claude Code into a full AI software team β 139 specialized agents that plan, build, review, test, and learn from every mistake. No configuration needed β just install and code.
> **v2.0**: 13 new agents (sast-scanner, mutation-tester, graph-analyst, mcp-manager, community-manager, benchmark, dependency-auditor, api-designer, incident-responder, data-modeler, test-architect, release-engineer, documentation-architect) + 23 new skills (SAST, compliance, product, marketing, MCP) + 4 new hooks + Agent Monitoring Dashboard + GitHub Actions CI/CD + MCP Auto-Discovery. See [UPGRADING.md](UPGRADING.md) for details.
> **v2.1**: 7 new skills (minimax-pdf, minimax-docx, minimax-xlsx, pptx-generator, frontend-dev, fullstack-dev, clone-website) + 2 new agents (document-generator, website-cloner). Document generation, pixel-perfect website cloning, and enhanced frontend/fullstack patterns.
> **v2.1.1**: 7 new skills from oh-my-claudecode (smart-model-routing, deep-interview, agent-benchmark, visual-verdict, ai-slop-cleaner, factcheck-guard, notepad-system) + 1 new rule (commit-trailers).
> **v2.2**: 5 features from Claude Code source β Agent Memory (persistent per-agent memory), Magic Docs (auto-updating docs), Dream Consolidation (cross-session memory cleanup), Smart Recall (frontmatter-based memory scoring), Plugin Toggle (hook enable/disable CLI). +7 hooks, skill references for 21 agents.
> **v2.2.1**: Monetization stack β 1 new agent (monetization-expert), 2 new skills (paywall-optimizer, codex-orchestration), 3 updated skills (revenuecat-patterns, paywall-strategy, subscription-pricing). AI-powered paywall optimization with 14-category benchmarks, RevenueCat SDK patterns, Codex + Claude Code orchestration.
> **v2.3**: `vibeco` CLI tool (stats, doctor, profiles, dashboard), 6 preset profiles for token savings, one-liner install (`curl | bash`).
> **v2.4**: Terminal HUD (real-time statusline), prompt auto-improver (enriches vague prompts with context), persistent planning system (PLAN.md/PROGRESS.md/CONTEXT.md for 96.7% task completion). Competitive gap closure from analysis of 20+ ecosystem repos.
> **v3.0**: `npx vibecosystem init` npm installer, `plugin.json` for official plugin ecosystem, worktree isolation on 60 producer agents, multi-LLM model routing (Haiku/Sonnet/Opus tiers), knowledge graph integration (6-71x token savings), dashboard v2 with token/cost tracking.
## The Problem
Claude Code is powerful, but it's one assistant. You prompt, it responds, you review. For complex projects you need a planner, a reviewer, a security auditor, a tester β and you end up being all of them yourself.
## The Solution
vibecosystem is a complete [Claude Code](https://docs.anthropic.com/en/docs/claude-code) ecosystem that creates a self-organizing AI team:
1. **139 agents** β specialized roles from frontend-dev to security-analyst
2. **295 skills** β reusable knowledge from TDD workflows to Kubernetes patterns
3. **73 hooks** β TypeScript sensors that observe, filter, and inject context
4. **20 rules** β behavioral guidelines that shape every agent's output
5. **Self-learning** β every error becomes a rule, automatically
After setup, you say "build a feature" and 20+ agents coordinate across 5 phases.
## Quick Start
### npm (recommended)
```bash
npx vibecosystem init
```
### One-liner
```bash
curl -fsSL https://raw.githubusercontent.com/vibeeval/vibecosystem/main/install-remote.sh | bash
```
### Manual
```bash
git clone https://github.com/vibeeval/vibecosystem.git
cd vibecosystem
./install.sh
```
That's it. Use Claude Code normally. The team activates.
### vibeco CLI
After install, the `vibeco` command is available:
```bash
vibeco stats # ecosystem statistics
vibeco list agents --search security # browse components
vibeco profile frontend # switch profile (saves tokens)
vibeco doctor # health check
vibeco dashboard # start monitoring UI
vibeco update # pull latest & reinstall
```
### Profiles
Save tokens by loading only what you need:
| Profile | Agents | Skills | Use case |
|---------|--------|--------|----------|
| `minimal` | ~15 | ~40 | Core only (review, test, verify) |
| `frontend` | ~30 | ~60 | React/Next.js/CSS/a11y |
| `backend` | ~44 | ~74 | API/DB/security |
| `fullstack` | ~59 | ~96 | Frontend + Backend |
| `devops` | ~33 | ~61 | CI/CD/K8s/cloud |
| `all` | 139 | 295 | Everything (default) |
```bash
vibeco profile frontend # switch to frontend profile
vibeco profile all # back to everything
```
## How It Works
```
YOU SAY SOMETHING VIBECOSYSTEM ACTIVATES RESULT
ββββββββββββββββ ββββββββββββββββββββββββ ββββββββββββ
β "add a new ββββ Intent βββ β Phase 1: scout + ββββ Code β Feature β
β feature" β Classifier β architect plan β Written β built, β
β β β Phase 2: backend-dev β Tested β reviewed,β
β β β + frontend-dev β Reviewedβ tested, β
β β β Phase 3: code-review β β merged β
β β β + security-review β β β
β β β Phase 4: verifier β β β
β β β Phase 5: self-learnerβ β β
ββββββββββββββββ ββββββββββββββββββββββββ ββββββββββββ
```
**Hooks** are sensors β they observe every tool call and inject relevant context:
```
"fix the bug" β compiler-in-loop + error-broadcast ~2,400 tok
"add api endpoint" β edit-context + signature-helper + arch ~3,100 tok
"explain this code" β (nothing extra) ~800 tok
```
**Agents** are muscles β each one specialized for a specific job:
```
GraphQL API β graphql-expert (backup: backend-dev)
Kubernetes β kubernetes-expert (backup: devops)
DDD modeling β ddd-expert (backup: architect)
Bug reproduction β replay (backup: sleuth)
... 70 more routing rules
```
**Self-Learning Pipeline** turns mistakes into permanent knowledge:
```
Error happens β passive-learner captures pattern (+ project tag)
β consolidator groups & counts (per-project + global)
β confidence >= 5 β auto-inject into context
β 2+ projects, 5+ total β cross-project promotion
β 10x repeat β permanent .md rule file
```
No manual intervention. The system writes its own rules β and shares them across projects.
## What's New in v2.0
- **SAST Security Scanner** β static analysis agent + hook for automated vulnerability detection
- **Agent Monitoring Dashboard** β real-time web UI for agent activity and performance
- **MCP Auto-Discovery** β automatic MCP server recommendations based on project type
- **Changelog Automation** β automatic changelog generation at session end
- **Compliance Skills** β SOC2, GDPR, HIPAA compliance checking
- **Product & Marketing Skills** β PRD writer, analytics setup, growth playbooks
- **GitHub Actions CI/CD** β automated PR review + issue fix workflows
- **Mutation Testing** β test quality measurement via mutation analysis
- **Code Knowledge Graph** β codebase structure analysis with graph-analyst
---
## Core Features
### Agent Swarm
Say "add a new feature" and 20+ agents activate across 5 phases.
```
Phase 1 (Discovery): scout + architect + project-manager
Phase 2 (Development): backend-dev + frontend-dev + devops + specialists
Phase 3 (Review): code-reviewer + security-reviewer + qa-engineer
Phase 4 (QA Loop): verifier + tdd-guide (max 3 retry β escalate)
Phase 5 (Final): self-learner + technical-writer
```
### Self-Learning Pipeline
Every error becomes a rule. Automatically.
### Dev-QA Loop
Every task goes through a quality gate:
```
Developer implements β code-reviewer + verifier check
β PASS β next task
β FAIL β feedback to developer, retry (max 3)
β 3x FAIL β escalate (reassign / decompose / defer)
```
### Cross-Project Learning
Patterns learned in one project automatically benefit all your projects.
```
Project A: add-error-handling (3x) ββ
ββ 2+ projects, 5+ total β GLOBAL
Project B: add-error-handling (4x) ββ
β
Next session in ANY project β "add-error-handling" injected as global pattern
```
Each project gets its own pattern store. When the same pattern appears in 2+ projects with 5+ total occurrences, it's promoted to a global pattern that benefits every project β even brand new ones.
```bash
node ~/.claude/hooks/dist/instinct-cli.mjs portfolio # All projects
node ~/.claude/hooks/dist/instinct-cli.mjs global # Global patterns
node ~/.claude/hooks/dist/instinct-cli.mjs project # Project detail
node ~/.claude/hooks/dist/instinct-cli.mjs stats # Statistics
```
### Canavar Cross-Training
When one agent makes a mistake, the entire team learns from it.
```
Agent error β error-ledger.jsonl β skill-matrix.json
β All agents get the lesson at session start
β Team-wide error prevention
```
### Adaptive Hook Loading
73 hooks exist but they don't all run at once. Intent determines which hooks fire.
---
## Architecture
```
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β Claude Code β
β β
β ββββββββββββ ββββββββββββ ββββββββββββ β
β β Hooks β β Agents β β Skills β β
β β (73) ββ β (139) ββ β (295) β β
β ββββββ¬ββββββ ββββββ¬ββββββ ββββββββββββ β
β β β β
β βΌ βΌ β
β ββββββββββββ ββββββββββββ β
β β Rules β β Memory β β
β β (20) β β (PgSQL) β β
β ββββββββββββ ββββββββββββ β
β β
β ββββββββββββββββββββββββββββββββββββββββ β
β β Self-Learning Pipeline β β
β β instincts β consolidate β rules β β
β β + cross-project promotion β β
β ββββββββββββββββββββββββββββββββββββββββ β
β β
β ββββββββββββββββββββββββββββββββββββββββ β
β β Canavar Cross-Training β β
β β error-ledger β skill-matrix β team β β
β ββββββββββββββββββββββββββββββββββββββββ β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
```
---
## Agent Categories
| Category | Count | Examples |
|----------|-------|---------|
| Core Dev | 14 | frontend-dev, backend-dev, kraken, spark, devops, browser-agent, website-cloner |
| Review & QA | 8 | code-reviewer, security-reviewer, verifier, qa-engineer |
| Domain Experts | 36 | graphql-expert, kubernetes-expert, ddd-expert, redis-expert, paywall-planner |
| Architecture | 8 | architect, planner, clean-arch-expert, cqrs-expert |
| Testing | 6 | tdd-guide, e2e-runner, arbiter, mocksmith |
| DevOps & Cloud | 12 | aws-expert, gcp-expert, azure-expert, terraform-expert |
| Analysis | 11 | scout, sleuth, data-analyst, profiler, strategist, harvest |
| Orchestration | 16 | nexus, sentinel, commander, neuron, vault, nitro |
| Documentation | 6 | technical-writer, doc-updater, copywriter, api-doc-generator, document-generator |
| Learning | 7 | self-learner, canavar, reputation-engine, session-replay-analyzer |
---
## Comparison
| Feature | vibecosystem | Single Claude Code | Cursor | aider |
|---------|:----------:|:------------------:|:------:|:-----:|
| Specialized agents | **139** | 0 | 0 | 0 |
| Self-learning | **Yes** | No | No | No |
| Agent swarm coordination | **Yes** | No | No | No |
| Cross-project learning | **Yes** | No | No | No |
| Cross-agent error training | **Yes** | No | No | No |
| Dev-QA retry loop | **Yes** | No | No | No |
| Adaptive hook loading | **Yes** | No | No | No |
| Assignment matrix routing | **Yes** | No | No | No |
| Claude Code native | **Yes** | Yes | No | No |
| Zero config after install | **Yes** | Yes | No | No |
---
## What's Included
| Component | Count | Description |
|-----------|-------|-------------|
| `agents/` | 139 | Markdown agent definitions with specialized prompts |
| `skills/` | 295 | Reusable knowledge β TDD, security, patterns, frameworks |
| `hooks/src/` | 73 | TypeScript hooks β sensors, learners, validators |
| `rules/` | 20 | Behavioral guidelines β coding style, safety, QA |
---
## Tech Stack
| Component | Technology |
|-----------|-----------|
| Runtime | Claude Code (Claude Max) |
| Models | Opus 4.6 / Sonnet 4.6 |
| Hook engine | TypeScript β esbuild β .mjs |
| Memory DB | PostgreSQL + pgvector (Docker) |
| Agent format | Markdown + YAML frontmatter |
| Skill format | prompt.md / SKILL.md |
| Cross-training | JSONL ledger + JSON skill matrix |
| Cross-project learning | Per-project instinct stores + global promotion |
---
## Philosophy
```
hooks are sensors. observe, filter, signal.
agents are muscles. build, produce, fix.
the bridge between them: context injection.
no direct RPC. no message passing. by design.
implicit coordination through context.
```
---
## Data & Privacy
- All data stays on your machine (`~/.claude/`)
- No network requests, no telemetry, no cloud sync
- Self-learned rules go to `~/.claude/rules/`
- Hooks run locally via Claude Code's native hook system
---
## Multi-CLI Support
vibecosystem works with multiple AI coding tools:
| CLI | Installer | Instructions File | What You Get |
|-----|-----------|-------------------|--------------|
| **Claude Code** | `./install.sh` | `CLAUDE.md` | Full support (agents + skills + hooks + rules) |
| **Cursor IDE** | `./install-cursor.sh` | `AGENTS.md` + `.cursor/rules/` | 6 MDC rules + AGENTS.md + skills |
| **Codex CLI** (OpenAI) | `./install-codex.sh` | `AGENTS.md` | Skills only (295 skills) |
| **OpenCode** | Manual | `AGENTS.md` | Skills only |
```bash
# For Cursor IDE users:
./install-cursor.sh /path/to/your/project
# For Codex CLI users:
./install-codex.sh
```
See [docs/codex-setup.md](docs/codex-setup.md) for Codex CLI setup, or copy `.cursor/rules/` into any Cursor project.
---
## Inspired By
vibecosystem stands on the shoulders of great open-source projects:
- **[Shannon](https://github.com/KeygraphHQ/Shannon)** by KeygraphHQ β Result pattern, pentest pipeline, comment philosophy
- **[UI UX Pro Max](https://github.com/nextlevelbuilder/ui-ux-pro-max)** by nextlevelbuilder β Named UX rules, UI style catalog, design token architecture
- **[Game Studios](https://github.com/Donchitos/game-studios)** by Donchitos β Context resilience, incremental writing, gate-check system
- **[Skill Gateway](https://github.com/buraksu42/skill-gateway)** by buraksu42 β Invisible skill routing, external catalog, one-question rule
- **[Pyxel](https://github.com/kitao/pyxel)** by kitao β Retro game engine patterns, pixel art constraints, MML audio
- **[copilot-orchestra](https://github.com/ShepAlderson/copilot-orchestra)** by ShepAlderson -- Phase-gated commits, plan documentation trail, 90% confidence threshold
- **[RevenueCat](https://www.revenuecat.com/)** -- Subscription infrastructure, category benchmarks, paywall patterns
- **[Trail of Bits Skills](https://github.com/trailofbits/skills)** by trailofbits -- Security audit patterns, variant analysis, false positive verification, sharp edges detection
---
## Contributing
Contributions welcome! Areas where help is needed:
- **More agent definitions** β specialized roles for your domain
- **More skill patterns** β framework-specific knowledge (Rails, Flutter, etc.)
- **Better hooks** β new sensors, smarter context injection
- **Documentation** β tutorials, guides, examples
- **Translations** β improve existing or add new languages
---
## Turkce
### Nedir?
vibecosystem, Claude Code'u tam bir AI yazilim ekibine donusturur. Tek bir asistan degil β planlayan, gelistiren, review yapan, test eden ve her hatasindan ogrenen **139 uzman agent'lik bir ekip**.
Ozel model yok. Ozel API yok. Sadece Claude Code'un hook + agent + rules sistemi, sonuna kadar kullanilmis.
### Hizli Baslangic
```bash
git clone https://github.com/vibeeval/vibecosystem.git
cd vibecosystem
./install.sh
```
### Nasil Calisir?
1. **Hook'lar sensor** β gozlemler, filtreler, isaret eder
2. **Agent'lar kas** β calisir, uretir, duzeltir
3. **Aralarindaki kopru:** context injection
4. **Direkt RPC yok** β bilerek boyle
5. **Context uzerinden implicit koordinasyon** calisiyor
### Felsefe
```
Kullanicinin hicbir sey hatirlamasina gerek yok.
Her sey otomatik.
```
---
## License
MIT
---
**Built by [@vibeeval](https://x.com/vibeeval)**
No custom model. No custom API. Just good engineering.