opspilot-ai 0.1.1__tar.gz → 0.1.3__tar.gz

{opspilot_ai-0.1.1 → opspilot_ai-0.1.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: opspilot-ai
-Version: 0.1.1
+Version: 0.1.3
 Summary: AI-powered production incident analysis tool using multi-agent systems
 Home-page: https://github.com/choudharikiranv15/OpsPilot-AI
 Author: Kiran Vijaykumar Choudhari
@@ -87,86 +87,196 @@ OpsPilot-AI uses a **multi-agent AI architecture** to understand your project's
 
 ---
 
-## 🚀 Quick Start
+## 🚀 Quick Start (2 Minutes Setup)
 
-### Prerequisites
+### Step 1: Install OpsPilot-AI
 
-- Python 3.8 or higher
-- At least ONE of the following LLM providers:
-  - [Ollama](https://ollama.ai/) (local, free, recommended)
-  - Google Gemini API key (free tier)
-  - OpenRouter API key (free models available)
-  - HuggingFace API token (free tier)
-
-### Installation
-
-**From PyPI (Recommended):**
 ```bash
-# Basic installation
 pip install opspilot-ai
+```
 
-# With Redis support (recommended for production)
-pip install opspilot-ai[redis]
+### Step 2: Setup LLM (Choose ONE Option)
 
-# With all integrations (Redis + AWS + Kubernetes)
-pip install opspilot-ai[all]
-```
+OpsPilot-AI needs an LLM to analyze your code. Choose **one** of these options:
 
-**From Source:**
-```bash
-# Clone repository
-git clone https://github.com/choudharikiranv15/OpsPilot-AI.git
-cd opspilot
+---
 
-# Install in development mode
-pip install -e ".[all]"
-```
+#### Option A: Ollama (Recommended - Free & Private)
 
-### LLM Setup
+Ollama runs locally on your machine. Your code never leaves your computer.
 
-**Option 1: Ollama (Recommended - Local & Free)**
+**For macOS/Linux:**
 ```bash
 # Install Ollama
 curl -fsSL https://ollama.ai/install.sh | sh
 
-# Pull Llama 3 model
+# Pull the AI model (one-time download, ~4GB)
 ollama pull llama3
+
+# Verify it's running
+ollama list
+```
+
+**For Windows:**
+1. Download Ollama from [ollama.ai/download](https://ollama.ai/download)
+2. Install and run the application
+3. Open terminal and run:
+```bash
+ollama pull llama3
+```
+
+---
+
+#### Option B: Cloud API (No Local Install Required)
+
+Use cloud-based LLMs with free tiers. Set **one** of these environment variables:
+
+**Google Gemini (Recommended Cloud Option):**
+```bash
+# Get free API key: https://makersuite.google.com/app/apikey
+export GOOGLE_API_KEY="your-api-key-here"
 ```
 
-**Option 2: Cloud Providers (FREE tiers)**
+**OpenRouter (100+ Models Available):**
 ```bash
-# Copy environment template
-cp .env.example .env
+# Get free API key: https://openrouter.ai/keys
+export OPENROUTER_API_KEY="your-api-key-here"
+```
 
-# Add your API keys to .env
-# GOOGLE_API_KEY=your-key-here
-# OPENROUTER_API_KEY=your-key-here
-# HUGGINGFACE_API_KEY=your-key-here
+**HuggingFace:**
+```bash
+# Get free token: https://huggingface.co/settings/tokens
+export HUGGINGFACE_API_KEY="your-api-key-here"
 ```
 
-See [FREE_LLM_SETUP.md](FREE_LLM_SETUP.md) for detailed setup instructions.
+**Windows Users (set environment variable):**
+```cmd
+set GOOGLE_API_KEY=your-api-key-here
+```
 
-### Usage
+---
 
-Navigate to your project directory and run:
+### Step 3: Analyze Your Project
 
 ```bash
-# Basic analysis
+# Navigate to your project
+cd /path/to/your/project
+
+# Run analysis
 opspilot analyze
+```
+
+That's it! OpsPilot-AI will analyze your project and provide diagnosis.
+
+---
 
+## 📖 Usage Examples
+
+### Basic Commands
+
+```bash
+# Quick analysis (fastest)
+opspilot analyze --mode quick
+
+# Deep analysis (thorough, recommended)
+opspilot analyze --mode deep
+
+# Verbose output (see what's happening)
+opspilot analyze --verbose
+```
+
+### Advanced Commands
+
+```bash
 # Analyze with production logs from S3
 opspilot analyze --log-source s3://my-bucket/logs/app.log
 
-# Analyze with deployment correlation
+# Analyze with deployment correlation (links errors to git commits)
 opspilot analyze --deployment-analysis --since-hours 48
 
-# JSON output for automation
+# JSON output for CI/CD automation
 opspilot analyze --json --mode quick
 
-# Verbose output for debugging
+# Analyze specific log file
+opspilot analyze --log-source /var/log/myapp/error.log
+
+# Full debugging output
 opspilot analyze --verbose --debug
 ```
 
+### Analysis Modes
+
+| Mode | Speed | LLM Calls | Use Case |
+|------|-------|-----------|----------|
+| `quick` | Fast | 1 | Quick check, CI/CD pipelines |
+| `deep` | Thorough | Up to 4 | Detailed incident analysis |
+| `explain` | Instant | 0 | Context gathering only (no LLM) |
+
+### Installation Options
+
+```bash
+# Basic (just the CLI)
+pip install opspilot-ai
+
+# With Redis support (remembers past incidents)
+pip install "opspilot-ai[redis]"
+
+# With AWS support (S3, CloudWatch logs)
+pip install "opspilot-ai[aws]"
+
+# With Kubernetes support (K8s pod logs)
+pip install "opspilot-ai[k8s]"
+
+# Everything included
+pip install "opspilot-ai[all]"
+```
+
+---
+
+## 🔧 Troubleshooting
+
+### "No LLM providers available"
+
+You need to set up an LLM. Choose one:
+
+```bash
+# Option 1: Install Ollama (recommended)
+curl -fsSL https://ollama.ai/install.sh | sh
+ollama pull llama3
+
+# Option 2: Set a cloud API key
+export GOOGLE_API_KEY="your-key"
+```
+
+### "Ollama connection refused"
+
+Make sure Ollama is running:
+```bash
+# Start Ollama service
+ollama serve
+
+# In another terminal, verify it works
+ollama list
+```
+
+### "No logs found"
+
+OpsPilot-AI looks for logs in these locations:
+- `./logs/` directory
+- `*.log` files in project root
+- Files specified with `--log-source`
+
+```bash
+# Specify log file directly
+opspilot analyze --log-source ./my-app/error.log
+```
+
+### "Context collected: 0 env vars"
+
+Make sure you have a `.env` file in your project directory, or environment variables set.
+
+---
+
 **Example Output:**
 
 ```

{opspilot_ai-0.1.1 → opspilot_ai-0.1.3}/README.md

@@ -28,86 +28,196 @@ OpsPilot-AI uses a **multi-agent AI architecture** to understand your project's
 
 ---
 
-## 🚀 Quick Start
+## 🚀 Quick Start (2 Minutes Setup)
 
-### Prerequisites
+### Step 1: Install OpsPilot-AI
 
-- Python 3.8 or higher
-- At least ONE of the following LLM providers:
-  - [Ollama](https://ollama.ai/) (local, free, recommended)
-  - Google Gemini API key (free tier)
-  - OpenRouter API key (free models available)
-  - HuggingFace API token (free tier)
-
-### Installation
-
-**From PyPI (Recommended):**
 ```bash
-# Basic installation
 pip install opspilot-ai
+```
 
-# With Redis support (recommended for production)
-pip install opspilot-ai[redis]
+### Step 2: Setup LLM (Choose ONE Option)
 
-# With all integrations (Redis + AWS + Kubernetes)
-pip install opspilot-ai[all]
-```
+OpsPilot-AI needs an LLM to analyze your code. Choose **one** of these options:
 
-**From Source:**
-```bash
-# Clone repository
-git clone https://github.com/choudharikiranv15/OpsPilot-AI.git
-cd opspilot
+---
 
-# Install in development mode
-pip install -e ".[all]"
-```
+#### Option A: Ollama (Recommended - Free & Private)
 
-### LLM Setup
+Ollama runs locally on your machine. Your code never leaves your computer.
 
-**Option 1: Ollama (Recommended - Local & Free)**
+**For macOS/Linux:**
 ```bash
 # Install Ollama
 curl -fsSL https://ollama.ai/install.sh | sh
 
-# Pull Llama 3 model
+# Pull the AI model (one-time download, ~4GB)
 ollama pull llama3
+
+# Verify it's running
+ollama list
+```
+
+**For Windows:**
+1. Download Ollama from [ollama.ai/download](https://ollama.ai/download)
+2. Install and run the application
+3. Open terminal and run:
+```bash
+ollama pull llama3
+```
+
+---
+
+#### Option B: Cloud API (No Local Install Required)
+
+Use cloud-based LLMs with free tiers. Set **one** of these environment variables:
+
+**Google Gemini (Recommended Cloud Option):**
+```bash
+# Get free API key: https://makersuite.google.com/app/apikey
+export GOOGLE_API_KEY="your-api-key-here"
 ```
 
-**Option 2: Cloud Providers (FREE tiers)**
+**OpenRouter (100+ Models Available):**
 ```bash
-# Copy environment template
-cp .env.example .env
+# Get free API key: https://openrouter.ai/keys
+export OPENROUTER_API_KEY="your-api-key-here"
+```
 
-# Add your API keys to .env
-# GOOGLE_API_KEY=your-key-here
-# OPENROUTER_API_KEY=your-key-here
-# HUGGINGFACE_API_KEY=your-key-here
+**HuggingFace:**
+```bash
+# Get free token: https://huggingface.co/settings/tokens
+export HUGGINGFACE_API_KEY="your-api-key-here"
 ```
 
-See [FREE_LLM_SETUP.md](FREE_LLM_SETUP.md) for detailed setup instructions.
+**Windows Users (set environment variable):**
+```cmd
+set GOOGLE_API_KEY=your-api-key-here
+```
 
-### Usage
+---
 
-Navigate to your project directory and run:
+### Step 3: Analyze Your Project
 
 ```bash
-# Basic analysis
+# Navigate to your project
+cd /path/to/your/project
+
+# Run analysis
 opspilot analyze
+```
+
+That's it! OpsPilot-AI will analyze your project and provide diagnosis.
+
+---
 
+## 📖 Usage Examples
+
+### Basic Commands
+
+```bash
+# Quick analysis (fastest)
+opspilot analyze --mode quick
+
+# Deep analysis (thorough, recommended)
+opspilot analyze --mode deep
+
+# Verbose output (see what's happening)
+opspilot analyze --verbose
+```
+
+### Advanced Commands
+
+```bash
 # Analyze with production logs from S3
 opspilot analyze --log-source s3://my-bucket/logs/app.log
 
-# Analyze with deployment correlation
+# Analyze with deployment correlation (links errors to git commits)
 opspilot analyze --deployment-analysis --since-hours 48
 
-# JSON output for automation
+# JSON output for CI/CD automation
 opspilot analyze --json --mode quick
 
-# Verbose output for debugging
+# Analyze specific log file
+opspilot analyze --log-source /var/log/myapp/error.log
+
+# Full debugging output
 opspilot analyze --verbose --debug
 ```
 
+### Analysis Modes
+
+| Mode | Speed | LLM Calls | Use Case |
+|------|-------|-----------|----------|
+| `quick` | Fast | 1 | Quick check, CI/CD pipelines |
+| `deep` | Thorough | Up to 4 | Detailed incident analysis |
+| `explain` | Instant | 0 | Context gathering only (no LLM) |
+
+### Installation Options
+
+```bash
+# Basic (just the CLI)
+pip install opspilot-ai
+
+# With Redis support (remembers past incidents)
+pip install "opspilot-ai[redis]"
+
+# With AWS support (S3, CloudWatch logs)
+pip install "opspilot-ai[aws]"
+
+# With Kubernetes support (K8s pod logs)
+pip install "opspilot-ai[k8s]"
+
+# Everything included
+pip install "opspilot-ai[all]"
+```
+
+---
+
+## 🔧 Troubleshooting
+
+### "No LLM providers available"
+
+You need to set up an LLM. Choose one:
+
+```bash
+# Option 1: Install Ollama (recommended)
+curl -fsSL https://ollama.ai/install.sh | sh
+ollama pull llama3
+
+# Option 2: Set a cloud API key
+export GOOGLE_API_KEY="your-key"
+```
+
+### "Ollama connection refused"
+
+Make sure Ollama is running:
+```bash
+# Start Ollama service
+ollama serve
+
+# In another terminal, verify it works
+ollama list
+```
+
+### "No logs found"
+
+OpsPilot-AI looks for logs in these locations:
+- `./logs/` directory
+- `*.log` files in project root
+- Files specified with `--log-source`
+
+```bash
+# Specify log file directly
+opspilot analyze --log-source ./my-app/error.log
+```
+
+### "Context collected: 0 env vars"
+
+Make sure you have a `.env` file in your project directory, or environment variables set.
+
+---
+
 **Example Output:**
 
 ```

{opspilot_ai-0.1.1 → opspilot_ai-0.1.3}/opspilot/cli.py

@@ -47,9 +47,26 @@ load_dotenv(verbose=False, override=False)  # Don't override existing env vars
 app = typer.Typer(help="OpsPilot - Agentic AI CLI for incident analysis")
 console = Console()
 
+__version__ = "0.1.2"
+
+
+def version_callback(value: bool):
+    if value:
+        console.print(f"OpsPilot version {__version__}")
+        raise typer.Exit()
+
 
 @app.callback()
-def main():
+def main(
+    version: bool = typer.Option(
+        None,
+        "--version",
+        "-V",
+        callback=version_callback,
+        is_eager=True,
+        help="Show OpsPilot version and exit"
+    )
+):
     """
     OpsPilot CLI entry point.
     """

{opspilot_ai-0.1.1 → opspilot_ai-0.1.3}/opspilot/utils/llm_providers.py

@@ -121,7 +121,7 @@ class OllamaProvider(LLMProvider):
 class OpenRouterProvider(LLMProvider):
     """OpenRouter provider - access to many free open-source models."""
 
-    def __init__(self, model: str = "google/gemini-2.0-flash-exp:free", timeout: int = 60):
+    def __init__(self, model: str = "openrouter/free", timeout: int = 60):
         super().__init__(timeout)
         self.model = model
         self.api_key = os.getenv("OPENROUTER_API_KEY")  # Free tier available
@@ -159,7 +159,15 @@ class OpenRouterProvider(LLMProvider):
                 json=payload,
                 timeout=self.timeout
             )
-            response.raise_for_status()
+
+            # Try to get more details from error response
+            if not response.ok:
+                try:
+                    error_data = response.json()
+                    error_msg = error_data.get("error", {}).get("message", response.text)
+                except Exception:
+                    error_msg = response.text or f"HTTP {response.status_code}"
+                raise RuntimeError(f"OpenRouter API error ({response.status_code}): {error_msg}")
 
             data = response.json()
             return data["choices"][0]["message"]["content"]

{opspilot_ai-0.1.1 → opspilot_ai-0.1.3}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "opspilot-ai"
-version = "0.1.1"
+version = "0.1.3"
 description = "AI-powered production incident analysis tool using multi-agent systems"
 readme = "README.md"
 authors = [