neuroncli 2.3.0__tar.gz → 3.1.0__tar.gz

neuroncli-3.1.0/PKG-INFO

@@ -0,0 +1,132 @@
+Metadata-Version: 2.4
+Name: neuroncli
+Version: 3.1.0
+Summary: AI Coding Agent — Free Claude Code alternative. Native Rust binary with plan mode, shell escapes, and Azure AI + OpenRouter support.
+Author-email: Rahul S <rahultech72216@gmail.com>
+License: MIT
+Project-URL: Homepage, https://zero-x.live
+Project-URL: Repository, https://github.com/RAHUL-DevelopeRR/neuroncli
+Project-URL: Issues, https://github.com/RAHUL-DevelopeRR/neuroncli/issues
+Keywords: ai,coding,agent,cli,llm,claude-code,rust,code-assistant,neuron
+Classifier: Development Status :: 4 - Beta
+Classifier: Environment :: Console
+Classifier: Intended Audience :: Developers
+Classifier: Topic :: Software Development
+Classifier: Topic :: Software Development :: Code Generators
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+Requires-Dist: requests>=2.28.0
+
+# NeuronCLI
+
+**Free AI Coding Agent** — Claude Code alternative. Native Rust binary with Azure AI + OpenRouter support.
+
+> Built by [zero-x Corporation](https://zero-x.live)
+
+---
+
+## Install
+
+```bash
+# Windows / macOS / Cloud VMs
+pip install neuroncli
+
+# Debian / Ubuntu (Python 3.11+)
+pip install neuroncli --break-system-packages
+
+# Or use pipx (recommended for Debian)
+pipx install neuroncli
+```
+
+> **v3.0.0** ships a native Rust binary (14MB) for maximum performance.  
+> On systems without the binary, the Python implementation runs as fallback.
+
+## Quick Start
+
+```bash
+# Start interactive REPL
+neuron
+
+# One-shot task
+neuron "fix the bug in main.py"
+
+# Plan mode — architecture first, execute later
+neuron --plan "add authentication to this API"
+
+# YOLO mode — no permission prompts
+neuron --yolo "refactor src/ to use async"
+```
+
+## Zero Setup — First Run
+
+1. Type `neuron` in any project directory
+2. Browser opens to **OpenRouter** login (free account)
+3. API key auto-provisioned + encrypted in `~/.neuroncli/vault.enc`
+4. **Done.** No manual config, no copy-paste.
+
+## Features
+
+| Feature | Description |
+|---------|-------------|
+| **Native Rust Binary** | 14MB single binary, instant startup |
+| **Plan Mode** | `/plan` generates architecture plans, `/plan off` to execute |
+| **? Shortcuts Panel** | Press `?` for Claude Code-style two-column cheatsheet |
+| **Shell Escape** | `!git status`, `!ls` — run commands without burning tokens |
+| **Dynamic Prompt** | Shows current mode: `>`, `[plan] >`, `[edit] >`, `[auto] >` |
+| **Azure AI Foundry** | GPT-5.5 primary (44K tokens/day free quota) |
+| **OpenRouter Fallback** | Free-tier model when Azure quota is exhausted |
+| **Directory Trust** | Interactive security prompt on first use per directory |
+| **Session Persistence** | Resume previous conversations with `/resume` |
+| **Git Auto-Commit** | Every AI edit is auto-committed |
+| **Context Compression** | `/compact` prevents context bloat |
+| **NEURON.md** | Project context file (like Claude's CLAUDE.md) |
+
+## Shortcuts (press `?` in REPL)
+
+| Key | Action | Key | Action |
+|-----|--------|-----|--------|
+| `?` | Shortcuts panel | `/help` | Full command list |
+| `!cmd` | Shell command | `/plan` | Toggle plan mode |
+| `Up/Down` | History | `/compact` | Compress context |
+| `Ctrl+C` | Cancel | `/clear` | Reset session |
+| `Ctrl+D` | Exit | `/model` | Switch model |
+
+## Plan Mode
+
+```
+> /plan                              ← Enter plan mode
+[plan] > add user authentication     ← Agent generates architecture plan
+[plan] > /plan off                   ← Exit plan mode
+> execute the plan                   ← Agent implements with full access
+```
+
+## CLI Flags
+
+| Flag | Description |
+|------|-------------|
+| `--yolo` | Skip all permission prompts |
+| `--plan` | Plan mode — reason before coding |
+| `--model`, `-m` | Model to use |
+| `--provider`, `-p` | `openrouter` or `ollama` |
+| `--dir`, `-d` | Working directory |
+| `--no-stream` | Disable token streaming |
+
+## Requirements
+
+- **Windows:** Works out of the box (native Rust binary)
+- **Linux/macOS:** Python 3.10+ (Python fallback), or install Rust binary separately
+- Internet connection for Azure AI / OpenRouter
+
+## License
+
+MIT — free to use, modify, and distribute commercially.
+
+---
+
+*By [zero-x Corporation](https://zero-x.live) — Tamil Nadu, India*

neuroncli-3.1.0/README.md

@@ -0,0 +1,107 @@
+# NeuronCLI
+
+**Free AI Coding Agent** — Claude Code alternative. Native Rust binary with Azure AI + OpenRouter support.
+
+> Built by [zero-x Corporation](https://zero-x.live)
+
+---
+
+## Install
+
+```bash
+# Windows / macOS / Cloud VMs
+pip install neuroncli
+
+# Debian / Ubuntu (Python 3.11+)
+pip install neuroncli --break-system-packages
+
+# Or use pipx (recommended for Debian)
+pipx install neuroncli
+```
+
+> **v3.0.0** ships a native Rust binary (14MB) for maximum performance.  
+> On systems without the binary, the Python implementation runs as fallback.
+
+## Quick Start
+
+```bash
+# Start interactive REPL
+neuron
+
+# One-shot task
+neuron "fix the bug in main.py"
+
+# Plan mode — architecture first, execute later
+neuron --plan "add authentication to this API"
+
+# YOLO mode — no permission prompts
+neuron --yolo "refactor src/ to use async"
+```
+
+## Zero Setup — First Run
+
+1. Type `neuron` in any project directory
+2. Browser opens to **OpenRouter** login (free account)
+3. API key auto-provisioned + encrypted in `~/.neuroncli/vault.enc`
+4. **Done.** No manual config, no copy-paste.
+
+## Features
+
+| Feature | Description |
+|---------|-------------|
+| **Native Rust Binary** | 14MB single binary, instant startup |
+| **Plan Mode** | `/plan` generates architecture plans, `/plan off` to execute |
+| **? Shortcuts Panel** | Press `?` for Claude Code-style two-column cheatsheet |
+| **Shell Escape** | `!git status`, `!ls` — run commands without burning tokens |
+| **Dynamic Prompt** | Shows current mode: `>`, `[plan] >`, `[edit] >`, `[auto] >` |
+| **Azure AI Foundry** | GPT-5.5 primary (44K tokens/day free quota) |
+| **OpenRouter Fallback** | Free-tier model when Azure quota is exhausted |
+| **Directory Trust** | Interactive security prompt on first use per directory |
+| **Session Persistence** | Resume previous conversations with `/resume` |
+| **Git Auto-Commit** | Every AI edit is auto-committed |
+| **Context Compression** | `/compact` prevents context bloat |
+| **NEURON.md** | Project context file (like Claude's CLAUDE.md) |
+
+## Shortcuts (press `?` in REPL)
+
+| Key | Action | Key | Action |
+|-----|--------|-----|--------|
+| `?` | Shortcuts panel | `/help` | Full command list |
+| `!cmd` | Shell command | `/plan` | Toggle plan mode |
+| `Up/Down` | History | `/compact` | Compress context |
+| `Ctrl+C` | Cancel | `/clear` | Reset session |
+| `Ctrl+D` | Exit | `/model` | Switch model |
+
+## Plan Mode
+
+```
+> /plan                              ← Enter plan mode
+[plan] > add user authentication     ← Agent generates architecture plan
+[plan] > /plan off                   ← Exit plan mode
+> execute the plan                   ← Agent implements with full access
+```
+
+## CLI Flags
+
+| Flag | Description |
+|------|-------------|
+| `--yolo` | Skip all permission prompts |
+| `--plan` | Plan mode — reason before coding |
+| `--model`, `-m` | Model to use |
+| `--provider`, `-p` | `openrouter` or `ollama` |
+| `--dir`, `-d` | Working directory |
+| `--no-stream` | Disable token streaming |
+
+## Requirements
+
+- **Windows:** Works out of the box (native Rust binary)
+- **Linux/macOS:** Python 3.10+ (Python fallback), or install Rust binary separately
+- Internet connection for Azure AI / OpenRouter
+
+## License
+
+MIT — free to use, modify, and distribute commercially.
+
+---
+
+*By [zero-x Corporation](https://zero-x.live) — Tamil Nadu, India*

{neuroncli-2.3.0 → neuroncli-3.1.0}/neuroncli/cli.py

@@ -353,6 +353,19 @@ def main(argv: list[str] | None = None) -> int:
         except Exception:
             pass
 
+    # ── Try native Rust binary first ─────────────────────────
+
+    native = _find_native_binary()
+    if native:
+        import subprocess
+        args_list = argv if argv is not None else sys.argv[1:]
+        try:
+            result = subprocess.run([native] + args_list)
+            return result.returncode
+        except (OSError, FileNotFoundError):
+            pass  # Fall through to Python implementation
+
+    # ── Python fallback ──────────────────────────────────────
     parser = build_parser()
     args = parser.parse_args(argv)
 
@@ -405,5 +418,35 @@ def main(argv: list[str] | None = None) -> int:
         return 0
 
 
+def _find_native_binary() -> str | None:
+    """Locate the native Rust binary — bundled in package or in PATH."""
+    import shutil
+
+    # Check bundled binary in package
+    pkg_dir = os.path.dirname(os.path.abspath(__file__))
+    ext = ".exe" if sys.platform == "win32" else ""
+    bundled = os.path.join(pkg_dir, "bin", f"neuron{ext}")
+    if os.path.isfile(bundled) and os.access(bundled, os.X_OK):
+        return bundled
+
+    # Check PATH (but not ourselves — avoid infinite recursion)
+    path_binary = shutil.which(f"neuron{ext}")
+    if path_binary:
+        # Verify it's the Rust binary by checking --version output
+        import subprocess
+        try:
+            result = subprocess.run(
+                [path_binary, "--version"],
+                capture_output=True, text=True, timeout=3
+            )
+            if "NeuronCLI" in result.stdout:
+                return path_binary
+        except (OSError, subprocess.TimeoutExpired):
+            pass
+
+    return None
+
+
 if __name__ == "__main__":
     raise SystemExit(main())
+