{"id":47661077,"url":"https://github.com/vibeeval/vibecosystem","last_synced_at":"2026-04-12T03:06:35.121Z","repository":{"id":345358093,"uuid":"1181624333","full_name":"vibeeval/vibecosystem","owner":"vibeeval","description":"Your AI software team. 137 agents, 269 skills, 53 hooks. Self-learning, multi-agent swarm, cross-project training. Built on Claude Code.","archived":false,"fork":false,"pushed_at":"2026-04-02T09:55:05.000Z","size":7549,"stargazers_count":373,"open_issues_count":5,"forks_count":29,"subscribers_count":8,"default_branch":"main","last_synced_at":"2026-04-02T23:58:13.136Z","etag":null,"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"],"latest_commit_sha":null,"homepage":null,"language":"C#","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/vibeeval.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":"SECURITY.md","support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null}},"created_at":"2026-03-14T11:57:55.000Z","updated_at":"2026-04-02T19:46:11.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/vibeeval/vibecosystem","commit_stats":null,"previous_names":["vibeeval/vibecosystem"],"tags_count":8,"template":false,"template_full_name":null,"purl":"pkg:github/vibeeval/vibecosystem","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vibeeval%2Fvibecosystem","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vibeeval%2Fvibecosystem/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vibeeval%2Fvibecosystem/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vibeeval%2Fvibecosystem/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/vibeeval","download_url":"https://codeload.github.com/vibeeval/vibecosystem/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vibeeval%2Fvibecosystem/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31421869,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-05T00:25:07.052Z","status":"online","status_checked_at":"2026-04-05T02:00:05.211Z","response_time":75,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["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"],"created_at":"2026-04-02T11:10:27.894Z","updated_at":"2026-04-12T03:06:35.114Z","avatar_url":"https://github.com/vibeeval.png","language":"C#","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cdiv align=\"center\"\u003e\n\n# vibecosystem\n\n**Your AI software team. Built on Claude Code.**\n\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)\n[![Agents](https://img.shields.io/badge/agents-139-blue.svg)](#agents)\n[![Skills](https://img.shields.io/badge/skills-295-green.svg)](#skills)\n[![Hooks](https://img.shields.io/badge/hooks-73-orange.svg)](#hooks)\n[![Rules](https://img.shields.io/badge/rules-20-red.svg)](#rules)\n[![Validate](https://github.com/vibeeval/vibecosystem/actions/workflows/validate.yml/badge.svg)](https://github.com/vibeeval/vibecosystem/actions/workflows/validate.yml)\n[![Works with Cursor](https://img.shields.io/badge/works%20with-Cursor-00b4d8.svg)](#multi-cli)\n[![Works with Codex CLI](https://img.shields.io/badge/works%20with-Codex%20CLI-10a37f.svg)](#multi-cli)\n[![Works with OpenCode](https://img.shields.io/badge/works%20with-OpenCode-purple.svg)](#multi-cli)\n\n[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) | [中文](docs/README_ZH.md) | [日本語](docs/README_JA.md) | [한국어](docs/README_KO.md) | [العربية](docs/README_AR.md) | [हिन्दी](docs/README_HI.md) | [Русский](docs/README_RU.md)\n\n![vibecosystem](assets/gif1-numbers.gif)\n\n\u003c/div\u003e\n\nvibecosystem turns Claude Code into a full AI software team — 139 specialized agents that plan, build, review, test, and learn from every mistake. No configuration needed — just install and code.\n\n\u003e **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.\n\n\u003e **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.\n\n\u003e **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).\n\n\u003e **v2.2**: 5 features from Claude Code source — 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.\n\n\u003e **v2.2.1**: Monetization stack — 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.\n\n\u003e **v2.3**: `vibeco` CLI tool (stats, doctor, profiles, dashboard), 6 preset profiles for token savings, one-liner install (`curl | bash`).\n\n\u003e **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.\n\n\u003e **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.\n\n## The Problem\n\nClaude 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 — and you end up being all of them yourself.\n\n## The Solution\n\nvibecosystem is a complete [Claude Code](https://docs.anthropic.com/en/docs/claude-code) ecosystem that creates a self-organizing AI team:\n\n1. \u003ca name=\"agents\"\u003e\u003c/a\u003e**139 agents** — specialized roles from frontend-dev to security-analyst\n2. \u003ca name=\"skills\"\u003e\u003c/a\u003e**295 skills** — reusable knowledge from TDD workflows to Kubernetes patterns\n3. \u003ca name=\"hooks\"\u003e\u003c/a\u003e**73 hooks** — TypeScript sensors that observe, filter, and inject context\n4. \u003ca name=\"rules\"\u003e\u003c/a\u003e**20 rules** — behavioral guidelines that shape every agent's output\n5. **Self-learning** — every error becomes a rule, automatically\n\nAfter setup, you say \"build a feature\" and 20+ agents coordinate across 5 phases.\n\n\u003ca name=\"english\"\u003e\u003c/a\u003e\n\n## Quick Start\n\n### npm (recommended)\n\n```bash\nnpx vibecosystem init\n```\n\n### One-liner\n\n```bash\ncurl -fsSL https://raw.githubusercontent.com/vibeeval/vibecosystem/main/install-remote.sh | bash\n```\n\n### Manual\n\n```bash\ngit clone https://github.com/vibeeval/vibecosystem.git\ncd vibecosystem\n./install.sh\n```\n\nThat's it. Use Claude Code normally. The team activates.\n\n### vibeco CLI\n\nAfter install, the `vibeco` command is available:\n\n```bash\nvibeco stats                          # ecosystem statistics\nvibeco list agents --search security  # browse components\nvibeco profile frontend               # switch profile (saves tokens)\nvibeco doctor                         # health check\nvibeco dashboard                      # start monitoring UI\nvibeco update                         # pull latest \u0026 reinstall\n```\n\n### Profiles\n\nSave tokens by loading only what you need:\n\n| Profile | Agents | Skills | Use case |\n|---------|--------|--------|----------|\n| `minimal` | ~15 | ~40 | Core only (review, test, verify) |\n| `frontend` | ~30 | ~60 | React/Next.js/CSS/a11y |\n| `backend` | ~44 | ~74 | API/DB/security |\n| `fullstack` | ~59 | ~96 | Frontend + Backend |\n| `devops` | ~33 | ~61 | CI/CD/K8s/cloud |\n| `all` | 139 | 295 | Everything (default) |\n\n```bash\nvibeco profile frontend  # switch to frontend profile\nvibeco profile all       # back to everything\n```\n\n## How It Works\n\n```\nYOU SAY SOMETHING                VIBECOSYSTEM ACTIVATES              RESULT\n┌──────────────┐                 ┌──────────────────────┐            ┌──────────┐\n│ \"add a new   │──→ Intent ──→  │ Phase 1: scout +     │──→ Code   │ Feature  │\n│  feature\"    │   Classifier   │   architect plan     │   Written │ built,   │\n│              │                 │ Phase 2: backend-dev │   Tested  │ reviewed,│\n│              │                 │   + frontend-dev     │   Reviewed│ tested,  │\n│              │                 │ Phase 3: code-review │           │ merged   │\n│              │                 │   + security-review  │           │          │\n│              │                 │ Phase 4: verifier    │           │          │\n│              │                 │ Phase 5: self-learner│           │          │\n└──────────────┘                 └──────────────────────┘            └──────────┘\n```\n\n**Hooks** are sensors — they observe every tool call and inject relevant context:\n```\n\"fix the bug\"       → compiler-in-loop + error-broadcast      ~2,400 tok\n\"add api endpoint\"  → edit-context + signature-helper + arch   ~3,100 tok\n\"explain this code\" → (nothing extra)                          ~800 tok\n```\n\n**Agents** are muscles — each one specialized for a specific job:\n```\nGraphQL API      → graphql-expert   (backup: backend-dev)\nKubernetes       → kubernetes-expert (backup: devops)\nDDD modeling     → ddd-expert       (backup: architect)\nBug reproduction → replay           (backup: sleuth)\n... 70 more routing rules\n```\n\n**Self-Learning Pipeline** turns mistakes into permanent knowledge:\n```\nError happens → passive-learner captures pattern (+ project tag)\n→ consolidator groups \u0026 counts (per-project + global)\n→ confidence \u003e= 5 → auto-inject into context\n→ 2+ projects, 5+ total → cross-project promotion\n→ 10x repeat → permanent .md rule file\n```\n\nNo manual intervention. The system writes its own rules — and shares them across projects.\n\n## What's New in v2.0\n\n- **SAST Security Scanner** — static analysis agent + hook for automated vulnerability detection\n- **Agent Monitoring Dashboard** — real-time web UI for agent activity and performance\n- **MCP Auto-Discovery** — automatic MCP server recommendations based on project type\n- **Changelog Automation** — automatic changelog generation at session end\n- **Compliance Skills** — SOC2, GDPR, HIPAA compliance checking\n- **Product \u0026 Marketing Skills** — PRD writer, analytics setup, growth playbooks\n- **GitHub Actions CI/CD** — automated PR review + issue fix workflows\n- **Mutation Testing** — test quality measurement via mutation analysis\n- **Code Knowledge Graph** — codebase structure analysis with graph-analyst\n\n---\n\n## Core Features\n\n### Agent Swarm\n\nSay \"add a new feature\" and 20+ agents activate across 5 phases.\n\n![Agent Swarm](assets/gif3-swarm.gif)\n\n```\nPhase 1 (Discovery):    scout + architect + project-manager\nPhase 2 (Development):  backend-dev + frontend-dev + devops + specialists\nPhase 3 (Review):       code-reviewer + security-reviewer + qa-engineer\nPhase 4 (QA Loop):      verifier + tdd-guide (max 3 retry → escalate)\nPhase 5 (Final):        self-learner + technical-writer\n```\n\n### Self-Learning Pipeline\n\nEvery error becomes a rule. Automatically.\n\n![Self-Learning](assets/gif2-pipeline.gif)\n\n### Dev-QA Loop\n\nEvery task goes through a quality gate:\n\n```\nDeveloper implements → code-reviewer + verifier check\n→ PASS → next task\n→ FAIL → feedback to developer, retry (max 3)\n→ 3x FAIL → escalate (reassign / decompose / defer)\n```\n\n### Cross-Project Learning\n\nPatterns learned in one project automatically benefit all your projects.\n\n```\nProject A: add-error-handling (3x) ─┐\n                                     ├→ 2+ projects, 5+ total → GLOBAL\nProject B: add-error-handling (4x) ─┘\n                                     ↓\nNext session in ANY project → \"add-error-handling\" injected as global pattern\n```\n\nEach 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 — even brand new ones.\n\n```bash\nnode ~/.claude/hooks/dist/instinct-cli.mjs portfolio      # All projects\nnode ~/.claude/hooks/dist/instinct-cli.mjs global          # Global patterns\nnode ~/.claude/hooks/dist/instinct-cli.mjs project \u003cname\u003e  # Project detail\nnode ~/.claude/hooks/dist/instinct-cli.mjs stats           # Statistics\n```\n\n### Canavar Cross-Training\n\nWhen one agent makes a mistake, the entire team learns from it.\n\n```\nAgent error → error-ledger.jsonl → skill-matrix.json\n→ All agents get the lesson at session start\n→ Team-wide error prevention\n```\n\n### Adaptive Hook Loading\n\n73 hooks exist but they don't all run at once. Intent determines which hooks fire.\n\n![Hooks](assets/gif4-hooks.gif)\n\n---\n\n## Architecture\n\n![Big Picture](assets/gif5-bigpicture.gif)\n\n```\n┌─────────────────────────────────────────────────────────┐\n│                    Claude Code                          │\n│                                                         │\n│  ┌──────────┐  ┌──────────┐  ┌──────────┐              │\n│  │  Hooks   │  │  Agents  │  │  Skills  │              │\n│  │  (73)    │→ │  (139)   │← │  (295)   │              │\n│  └────┬─────┘  └────┬─────┘  └──────────┘              │\n│       │              │                                   │\n│       ▼              ▼                                   │\n│  ┌──────────┐  ┌──────────┐                              │\n│  │  Rules   │  │  Memory  │                              │\n│  │  (20)    │  │ (PgSQL)  │                              │\n│  └──────────┘  └──────────┘                              │\n│                                                         │\n│  ┌──────────────────────────────────────┐                │\n│  │  Self-Learning Pipeline             │                │\n│  │  instincts → consolidate → rules    │                │\n│  │  + cross-project promotion          │                │\n│  └──────────────────────────────────────┘                │\n│                                                         │\n│  ┌──────────────────────────────────────┐                │\n│  │  Canavar Cross-Training             │                │\n│  │  error-ledger → skill-matrix → team │                │\n│  └──────────────────────────────────────┘                │\n└─────────────────────────────────────────────────────────┘\n```\n\n---\n\n## Agent Categories\n\n| Category | Count | Examples |\n|----------|-------|---------|\n| Core Dev | 14 | frontend-dev, backend-dev, kraken, spark, devops, browser-agent, website-cloner |\n| Review \u0026 QA | 8 | code-reviewer, security-reviewer, verifier, qa-engineer |\n| Domain Experts | 36 | graphql-expert, kubernetes-expert, ddd-expert, redis-expert, paywall-planner |\n| Architecture | 8 | architect, planner, clean-arch-expert, cqrs-expert |\n| Testing | 6 | tdd-guide, e2e-runner, arbiter, mocksmith |\n| DevOps \u0026 Cloud | 12 | aws-expert, gcp-expert, azure-expert, terraform-expert |\n| Analysis | 11 | scout, sleuth, data-analyst, profiler, strategist, harvest |\n| Orchestration | 16 | nexus, sentinel, commander, neuron, vault, nitro |\n| Documentation | 6 | technical-writer, doc-updater, copywriter, api-doc-generator, document-generator |\n| Learning | 7 | self-learner, canavar, reputation-engine, session-replay-analyzer |\n\n---\n\n## Comparison\n\n| Feature | vibecosystem | Single Claude Code | Cursor | aider |\n|---------|:----------:|:------------------:|:------:|:-----:|\n| Specialized agents | **139** | 0 | 0 | 0 |\n| Self-learning | **Yes** | No | No | No |\n| Agent swarm coordination | **Yes** | No | No | No |\n| Cross-project learning | **Yes** | No | No | No |\n| Cross-agent error training | **Yes** | No | No | No |\n| Dev-QA retry loop | **Yes** | No | No | No |\n| Adaptive hook loading | **Yes** | No | No | No |\n| Assignment matrix routing | **Yes** | No | No | No |\n| Claude Code native | **Yes** | Yes | No | No |\n| Zero config after install | **Yes** | Yes | No | No |\n\n---\n\n## What's Included\n\n| Component | Count | Description |\n|-----------|-------|-------------|\n| `agents/` | 139 | Markdown agent definitions with specialized prompts |\n| `skills/` | 295 | Reusable knowledge — TDD, security, patterns, frameworks |\n| `hooks/src/` | 73 | TypeScript hooks — sensors, learners, validators |\n| `rules/` | 20 | Behavioral guidelines — coding style, safety, QA |\n\n---\n\n## Tech Stack\n\n| Component | Technology |\n|-----------|-----------|\n| Runtime | Claude Code (Claude Max) |\n| Models | Opus 4.6 / Sonnet 4.6 |\n| Hook engine | TypeScript → esbuild → .mjs |\n| Memory DB | PostgreSQL + pgvector (Docker) |\n| Agent format | Markdown + YAML frontmatter |\n| Skill format | prompt.md / SKILL.md |\n| Cross-training | JSONL ledger + JSON skill matrix |\n| Cross-project learning | Per-project instinct stores + global promotion |\n\n---\n\n## Philosophy\n\n```\nhooks are sensors. observe, filter, signal.\nagents are muscles. build, produce, fix.\nthe bridge between them: context injection.\nno direct RPC. no message passing. by design.\nimplicit coordination through context.\n```\n\n---\n\n## Data \u0026 Privacy\n\n- All data stays on your machine (`~/.claude/`)\n- No network requests, no telemetry, no cloud sync\n- Self-learned rules go to `~/.claude/rules/`\n- Hooks run locally via Claude Code's native hook system\n\n---\n\n\u003ca name=\"multi-cli\"\u003e\u003c/a\u003e\n\n## Multi-CLI Support\n\nvibecosystem works with multiple AI coding tools:\n\n| CLI | Installer | Instructions File | What You Get |\n|-----|-----------|-------------------|--------------|\n| **Claude Code** | `./install.sh` | `CLAUDE.md` | Full support (agents + skills + hooks + rules) |\n| **Cursor IDE** | `./install-cursor.sh` | `AGENTS.md` + `.cursor/rules/` | 6 MDC rules + AGENTS.md + skills |\n| **Codex CLI** (OpenAI) | `./install-codex.sh` | `AGENTS.md` | Skills only (295 skills) |\n| **OpenCode** | Manual | `AGENTS.md` | Skills only |\n\n```bash\n# For Cursor IDE users:\n./install-cursor.sh /path/to/your/project\n\n# For Codex CLI users:\n./install-codex.sh\n```\n\nSee [docs/codex-setup.md](docs/codex-setup.md) for Codex CLI setup, or copy `.cursor/rules/` into any Cursor project.\n\n---\n\n## Inspired By\n\nvibecosystem stands on the shoulders of great open-source projects:\n\n- **[Shannon](https://github.com/KeygraphHQ/Shannon)** by KeygraphHQ — Result\u003cT,E\u003e pattern, pentest pipeline, comment philosophy\n- **[UI UX Pro Max](https://github.com/nextlevelbuilder/ui-ux-pro-max)** by nextlevelbuilder — Named UX rules, UI style catalog, design token architecture\n- **[Game Studios](https://github.com/Donchitos/game-studios)** by Donchitos — Context resilience, incremental writing, gate-check system\n- **[Skill Gateway](https://github.com/buraksu42/skill-gateway)** by buraksu42 — Invisible skill routing, external catalog, one-question rule\n- **[Pyxel](https://github.com/kitao/pyxel)** by kitao — Retro game engine patterns, pixel art constraints, MML audio\n- **[copilot-orchestra](https://github.com/ShepAlderson/copilot-orchestra)** by ShepAlderson -- Phase-gated commits, plan documentation trail, 90% confidence threshold\n- **[RevenueCat](https://www.revenuecat.com/)** -- Subscription infrastructure, category benchmarks, paywall patterns\n- **[Trail of Bits Skills](https://github.com/trailofbits/skills)** by trailofbits -- Security audit patterns, variant analysis, false positive verification, sharp edges detection\n\n---\n\n## Contributing\n\nContributions welcome! Areas where help is needed:\n\n- **More agent definitions** — specialized roles for your domain\n- **More skill patterns** — framework-specific knowledge (Rails, Flutter, etc.)\n- **Better hooks** — new sensors, smarter context injection\n- **Documentation** — tutorials, guides, examples\n- **Translations** — improve existing or add new languages\n\n---\n\n\u003ca name=\"turkce\"\u003e\u003c/a\u003e\n\n## Turkce\n\n### Nedir?\n\nvibecosystem, Claude Code'u tam bir AI yazilim ekibine donusturur. Tek bir asistan degil — planlayan, gelistiren, review yapan, test eden ve her hatasindan ogrenen **139 uzman agent'lik bir ekip**.\n\nOzel model yok. Ozel API yok. Sadece Claude Code'un hook + agent + rules sistemi, sonuna kadar kullanilmis.\n\n### Hizli Baslangic\n\n```bash\ngit clone https://github.com/vibeeval/vibecosystem.git\ncd vibecosystem\n./install.sh\n```\n\n### Nasil Calisir?\n\n1. **Hook'lar sensor** — gozlemler, filtreler, isaret eder\n2. **Agent'lar kas** — calisir, uretir, duzeltir\n3. **Aralarindaki kopru:** context injection\n4. **Direkt RPC yok** — bilerek boyle\n5. **Context uzerinden implicit koordinasyon** calisiyor\n\n### Felsefe\n\n```\nKullanicinin hicbir sey hatirlamasina gerek yok.\nHer sey otomatik.\n```\n\n---\n\n## License\n\nMIT\n\n---\n\n\u003cdiv align=\"center\"\u003e\n\n**Built by [@vibeeval](https://x.com/vibeeval)**\n\nNo custom model. No custom API. Just good engineering.\n\n\u003c/div\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvibeeval%2Fvibecosystem","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fvibeeval%2Fvibecosystem","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvibeeval%2Fvibecosystem/lists"}