⚡ FAZAH AI Arsenal v6.5 (github-ai-arsenal)
Autonomous AI Tool Broker & 17-Tool Native MCP Server (+ Prompts & Resources) — 20,078+ GitHub Tools, 1-Call Delivery (fazah g), Code Snippet Extractor, Isolated Sandbox, Local Web UI & Security Scanner
FAZAH AI Arsenal (
github-ai-arsenal) is a zero-dependency Universal Agent Skill (SKILL.md), 12-Tool Native Model Context Protocol (MCP) Server, and ultra-shortfazahCLI that equips Claude Code, Cursor, Google Antigravity, DeepSeek Harness, Windsurf, and OpenAI Codex with instant offline & online access to 20,078+ real, deduplicated GitHub repositories across 37 engineering domains.
👨💻 Author & Official Contact
| Field | Details |
|---|---|
| Creator & Lead Developer | FAZAH |
| Telegram | @testtttttttt11222 |
| mhamadabdalla150150@gmail.com | |
| Official Repository | github.com/fazah-arsenal/github-ai-arsenal |
🌟 Why FAZAH AI Arsenal v6.5?
Instead of letting your AI agent reinvent existing tools from scratch or burn through GitHub API rate limits, FAZAH AI Arsenal bundles a compressed offline index of 20,078+ real, battle-tested GitHub repositories alongside a pure-Python BM25 search engine, Repository Trust & Security Auditor, Side-by-Side Comparison Engine, and 1-Command Project Stack Scaffolder — with zero external Python dependencies.
✨ Key Capabilities:
- 🔌 12-Tool Native MCP Server (
scripts/mcp_server.py): Plugs directly into Antigravity, Claude Code, Cursor, Windsurf, and DeepSeek via standardstdio JSON-RPC 2.0— zeropip installrequired. - ⚡ Instant Offline BM25 Search (
fazah m&fazah r): Search 20,078+ real GitHub repositories across 37 domains in milliseconds with zero API rate limits (supports both English & Arabic prompts). - 🛡️ Security & Trust Auditor (
fazah a&fazah cmp): Computes a Trust Score (0–100) checking stars, maintenance recency, SPDX license, and repository hygiene — plus side-by-side repo comparisons. - 🏗️ 1-Command Project Scaffolder (
fazah st <stack> -s <dir>): Generates ready-to-run starter projects from 8 turnkey blueprints (saas-ai,telegram-ai-bot,scraping-rag,security-pentest-lab,reverse-engineering-lab,desktop-app,voice-ai-agent,mcp-builder). - 🚀 Smart Quickstart (
fazah q): Auto-detects any repository's ecosystem (Python, Node, Rust, Go, Docker) and outputs exact install/run commands. - 💉 Auto MCP Injector (
fazah mi): Safely injects any MCP server into.mcp.jsonwith automatic.bakbackup. - ⭐ Local Bookmarks (
fazah bm): Save favorite repositories with custom notes for instant recall. - 📦 Pull & Install Without Git (
fazah p&fazah is): Download any GitHub repo ZIP or install any repo as a localSKILL.mdwithout requiringgit.
🖥️ Live CLI Preview
⚡ 1-Command Installation
Installs the Universal Skill into ~/.agents/skills, ~/.gemini/config/skills (plus Claude/Cursor/Codex if present) and registers the global fazah CLI command automatically:
Windows (PowerShell)
irm "https://raw.githubusercontent.com/fazah-arsenal/github-ai-arsenal/main/install.ps1" | iex
(Or locally after downloading: .\install.ps1)
Linux / macOS
curl -fsSL "https://raw.githubusercontent.com/fazah-arsenal/github-ai-arsenal/main/install.sh" | bash
(Or locally: chmod +x install.sh && ./install.sh)
🔌 Connect as a Native MCP Server
Inject automatically into your workspace .mcp.json:
fazah mi github-ai-arsenal
Or paste into your MCP client configuration (Antigravity, Claude Code, Cursor, Windsurf, DeepSeek):
{
"mcpServers": {
"github-ai-arsenal": {
"command": "python",
"args": ["~/.agents/skills/github-ai-arsenal/scripts/mcp_server.py"]
}
}
}
🧰 17 Native MCP Tools (+ 4 Resources & 3 Prompts) Exposed:
| MCP Tool | Description |
|---|---|
arsenal_give_tool |
1-Call Autonomous Tool Broker: picks #1 tool, scans security, extracts code snippet & optionally pulls |
arsenal_get_snippet |
Extracts clean runnable code snippets from any README (saves 85% AI tokens) |
arsenal_deep_scan |
Deep supply-chain security scan of setup.py, package.json, install.sh, Dockerfile |
arsenal_analyze_project |
Scans local project directory & recommends missing GitHub tools & MCP servers |
arsenal_memory |
Persistent agent experience memory for verified working tools on this machine |
arsenal_recommend |
Intelligent tool recommender (Core Elite + 20,078+ Mega Index) |
arsenal_search_offline |
Instant offline BM25 search across 20,078+ repos with zero rate limits |
arsenal_stack |
Browse 8 turnkey project stacks with install commands |
arsenal_scaffold_stack |
Scaffold a complete starter project folder (README.md, setup_stack, .env.example, .mcp.json) |
arsenal_quickstart |
Auto-detect language/ecosystem and output exact install & run commands |
arsenal_compare |
Compare 2+ GitHub repositories side-by-side (Trust Score, Stars, License) |
arsenal_audit_repo |
Security & Trust Score (0–100) audit before pulling or running code |
arsenal_bookmark |
Add, list, or remove bookmarked favorite tools with notes |
arsenal_mcp_inject |
Safely inject any MCP server into .mcp.json with backup |
arsenal_inspect_repo |
Inspect repo metadata, top files, and README.md/SKILL.md without cloning |
arsenal_pull_repo |
Download and extract any GitHub repository ZIP without git |
arsenal_install_skill |
Auto-discover or synthesize SKILL.md and install any repo as a local skill |
🛠️ Ultra-Short CLI Cheatsheet (fazah)
fazah g "convert pdf to markdown" # 1-Call Autonomous Tool Broker (pick + scan + snippet)
fazah g "web scraper" --pull # Pick, security-scan & auto-download ZIP in 1 call
fazah sn unclecode/crawl4ai # Extract clean runnable code snippets (saves 85% tokens)
fazah mem log unclecode/crawl4ai -t scrape # Log verified working tool into agent experience memory
fazah try unclecode/crawl4ai # Pull into isolated Python venv sandbox safely
fazah ui # Launch Interactive Local Web Dashboard (http://127.0.0.1:8765)
fazah az . # Analyze current project & recommend missing tools/MCPs
fazah scan unclecode/crawl4ai # Deep supply-chain security scan of setup.py/package.json
fazah doc # Check local runtimes, skills & index health
fazah info # Show breakdown of 20,078+ tools across 37 domains
fazah r "telegram bot and web scraping" # Smart tool recommender (English & Arabic)
fazah m -q "postgres vector" -n 10 # Instant offline BM25 search across 20k+ repos
fazah q unclecode/crawl4ai # Auto-detect install & run commands
fazah a modelcontextprotocol/servers # Security & Trust Score (0-100) audit
fazah cmp unclecode/crawl4ai mendableai/firecrawl # Side-by-side repo comparison
fazah st list # List all 8 turnkey project stacks
fazah st saas-ai -s ./my-saas-app # Scaffold a complete starter project
fazah bm add unclecode/crawl4ai -m "Best" # Bookmark a favorite repository
fazah mi github-ai-arsenal # Auto-inject MCP server into .mcp.json
fazah t -t mcp # Fetch live trending GitHub repos
fazah s "deepseek coding agent" # Live GitHub API search
fazah i danielmiessler/fabric # Inspect repo structure & README without cloning
fazah p unclecode/crawl4ai -d ./tools # Download & extract repo ZIP without git
fazah is anthropics/skills # Install any GitHub repo as a local SKILL.md
fazah up # Self-update FAZAH AI Arsenal & 20k+ index from GitHub
📦 Install via Python pip
pip install git+https://github.com/fazah-arsenal/github-ai-arsenal.git
(Full command names like fazah recommend and python scripts/gh_arsenal.py recommend also work identically.)
📊 Domain Coverage (20,078+ Real Indexed Repositories)
| Domain | Indexed Tools | Highlights |
|---|---|---|
mcp-servers |
5,032 | Official & community MCP servers for DBs, browsers, cloud, filesystems |
go-systems & rust-systems |
4,613 | High-performance systems, networking, CLI & infrastructure tools |
devops-cloud & cpp-systems |
2,289 | Docker, Kubernetes, CI/CD, cloud-native & low-level C/C++ libraries |
machine-learning & ai-agents & ai-llm |
1,452 | Multi-agent swarms, LLM routers, local inference, RAG & fine-tuning |
ui-frontend & javascript-web |
1,418 | React, Next.js, Tailwind, shadcn/ui & modern frontend ecosystems |
media-audio-video |
1,130 | Speech-to-Text (Whisper), TTS (Kokoro), video processing & FFmpeg tools |
security-pentest & osint-recon & reverse-engineering |
778 | SAST scanners, secret hunters, APK/ASAR/binary decompilers & OSINT |
dotnet-csharp / python / java-jvm / php-web / mobile-desktop / blockchain-web3 |
2,650+ | Full-stack libraries, desktop GUIs, mobile apps, Web3 & game dev |
🧪 Running Automated Tests
python -m unittest tests/test_arsenal.py
📞 Connect With the Developer
- Author: FAZAH
- Telegram: @testtttttttt11222
- Email:
mhamadabdalla150150@gmail.com
📄 License
Released under the MIT License — free for personal, open-source, and commercial AI workflows.
Metadata
Release files for fazah-arsenal 6.5.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| fazah_arsenal-6.5.0.tar.gz | 885.4 kB | Without provenance | Details for fazah_arsenal-6.5.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| fazah_arsenal-6.5.0-py3-none-any.whl | Python 3 | none | any | Details for fazah_arsenal-6.5.0-py3-none-any.whl |
Total release size: 1.8 MB
Release files / fazah_arsenal-6.5.0.tar.gz
| Download URL | fazah_arsenal-6.5.0.tar.gz |
|---|---|
| Size | 885.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
bf628af507294dd248f57bf39d9db297cd3306ba3fc947b1c45b3079c9d22add
|
|
BLAKE2b-256 checksum How to use checksums |
76d7b1749fdde6fcbf50cb94005e3681bb7bc775779577f77ab2b9392f3af8e4
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.12.10
|
Release files / fazah_arsenal-6.5.0-py3-none-any.whl
| Download URL | fazah_arsenal-6.5.0-py3-none-any.whl |
|---|---|
| Size | 885.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
9332be362d2b25bc589f3821ec24b8d0dab1358f9a19f534271e6d3f65596b46
|
|
BLAKE2b-256 checksum How to use checksums |
3d577159d2b9b7b29446d9edc666701f624b4657a11aa3d511776b6c6c40a3a6
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.12.10
|
Security for fazah-arsenal 6.5.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.