Skip to content

Commit f2ef54b

Browse files
pavanjavapavanjava
authored
modified readme and feature release notes (#20)
1 parent 6cdc5b9 commit f2ef54b

2 files changed

Lines changed: 121 additions & 50 deletions

File tree

‎README.md‎

Lines changed: 103 additions & 50 deletions
Original file line numberDiff line numberDiff line change
@@ -7,7 +7,14 @@ An AI-powered CLI tool to assist with DevOps troubleshooting, Applications with
77
- 📊 **Log Analysis**: Analyze log files and get actionable insights
88
- 💬 **Query Interface**: Ask questions about DevOps best practices, Terraform, Kubernetes, etc.
99
- 🛠️ **Template Generation**: Generate infrastructure code templates
10-
- 🤖 **AI-Powered**: Leverages Claude AI for intelligent responses
10+
- 🤖 **AI-Powered**: Leverages multiple LLM providers (OpenAI, Anthropic, Gemini, Ollama, vLLM)
11+
- 🎯 **Flexible Provider Selection**: Choose your preferred LLM provider and model dynamically
12+
- 🔒 **Self-Hosted Options**: Run privately with Ollama or vLLM
13+
- 🧠 **Reasoning Mode**: Enable advanced reasoning capabilities for complex queries
14+
- 🐛 **Debug Mode**: Troubleshoot agent behavior with detailed logging
15+
- 💾 **Memory Management**: Persistent context using Qdrant vector database
16+
- 🎨 **Interactive Mode**: Engage in continuous conversations with the agent
17+
- 📝 **Multiple Output Formats**: Export results as text, JSON, or Markdown
1118

1219
## Installation
1320

@@ -25,63 +32,130 @@ pip install devops-agent
2532

2633
## Configuration
2734
#### LLM API KEYS
28-
```env
29-
export GEMINI_API_KEY=YOUR API KEY
30-
or
31-
export ANTHROPIC_API_KEY=YOUR API KEY
32-
or
33-
export OPENAI_API_KEY=YOUR API KEY
35+
```bash
36+
# For OpenAI
37+
export OPENAI_API_KEY=YOUR_API_KEY
38+
39+
# For Anthropic Claude
40+
export ANTHROPIC_API_KEY=YOUR_API_KEY
41+
42+
# For Google Gemini
43+
export GEMINI_API_KEY=YOUR_API_KEY
44+
45+
# For Ollama (self-hosted, typically no API key needed)
46+
export OLLAMA_API_KEY=YOUR_API_KEY # Optional
47+
48+
# For vLLM (self-hosted)
49+
export VLLM_API_KEY=YOUR_API_KEY
3450
```
3551
#### Qdrant Config for Agent Memory
52+
(If not configured fall backs to in-memory vector store)
3653
```env
3754
export QDRANT_URL=YOUR QDRANT URL
3855
export QDRANT_API_KEY=YOUR QDRANT API KEY
3956
```
4057
## Usage
4158

42-
### Analyze Log Files
59+
#### Ask Questions
4360

4461
```bash
45-
devops-agent run --log-file /path/to/app.log
62+
devops-agent run --query "I need terraform script to spin up Azure blob storage"
63+
devops-agent run --query "How to increase my pod memory and CPU in k8s"
4664
```
4765

48-
### Ask Questions
66+
#### Interactive Mode
4967

5068
```bash
51-
devops-agent run --query "I need terraform script to spin up Azure blob storage"
52-
devops-agent run --query "How to increase my pod memory and CPU in k8s"
69+
devops-agent run --interactive
70+
# or
71+
devops-agent run -i
5372
```
5473

55-
### Generate Templates
74+
### Advanced Options
75+
76+
#### Choose Your LLM Provider and Model
5677

5778
```bash
58-
devops-agent template terraform
59-
devops-agent template kubernetes
60-
devops-agent template docker
79+
# Use OpenAI with a specific model
80+
devops-agent run --provider openai --model gpt-4o --query "your question"
81+
82+
# Use Anthropic Claude
83+
devops-agent run --provider anthropic --model claude-sonnet-4-20250514 --query "your question"
84+
85+
# Use Google Gemini
86+
devops-agent run --provider google --model gemini-2.0-flash-exp --query "your question"
87+
88+
# Use Ollama (self-hosted)
89+
devops-agent run --provider ollama --model llama3 --query "your question"
90+
91+
# Use vLLM (self-hosted)
92+
devops-agent run --provider vllm --model your-model-name --query "your question"
6193
```
6294

63-
### Configuration
95+
#### Enable Debug Mode
6496

6597
```bash
66-
devops-agent config
98+
devops-agent run --query "your question" --debug_mode true
6799
```
68100

69-
## Examples
101+
#### Enable Reasoning Mode
70102

71103
```bash
72-
# Analyze application logs
73-
devops-agent run --log-file ./logs/app.log --format json
104+
devops-agent run --query "your question" --reasoning_enabled true
105+
```
74106

75-
# Get Terraform help
76-
devops-agent run --query "terraform script for AWS S3 bucket with versioning"
107+
#### Combine Multiple Options
77108

78-
# Kubernetes troubleshooting
79-
devops-agent run --query "pod is in CrashLoopBackOff status, how to debug?"
109+
```bash
110+
# Interactive mode with specific provider, model, and reasoning
111+
devops-agent run -i --provider anthropic --model claude-sonnet-4-20250514 --reasoning_enabled true
80112

81-
# Save output to file
82-
devops-agent run --query "docker-compose for nginx and postgres" --output docker-compose.yml
113+
# Query with debug mode and custom output
114+
devops-agent run --query "docker setup for microservices" --provider openai --model gpt-4o --debug_mode true --output result.md --format markdown
83115
```
84116

117+
## CLI Options Reference
118+
119+
### `devops-agent run` Options
120+
121+
| Option | Type | Description |
122+
|--------|------|-------------|
123+
| `--log-file` | Path | Path to log file to analyze |
124+
| `--provider` | String | LLM provider (openai, anthropic, google, ollama, vllm) |
125+
| `--model` | String | Model name (e.g., gpt-4o, claude-sonnet-4-20250514, gemini-2.0-flash-exp) |
126+
| `--query` | String | Query to ask the DevOps agent |
127+
| `--output` | Path | Output file path for saving results |
128+
| `--format` | Choice | Output format: text, json, or markdown (default: text) |
129+
| `--interactive, -i` | Flag | Run in interactive mode for continuous conversation |
130+
| `--debug_mode` | Boolean | Enable debug mode with detailed logging |
131+
| `--reasoning_enabled` | Boolean | Enable reasoning mode for complex problem-solving |
132+
133+
### Provider-Specific Model Examples
134+
135+
**OpenAI:**
136+
- `gpt-4o`
137+
- `gpt-5-mini`
138+
- `gpt-5.1`
139+
140+
**Anthropic:**
141+
- `claude-sonnet-4-20250514`
142+
- `claude-sonnet-4-5-20250929`
143+
- `claude-3-5-sonnet-20241022`
144+
145+
**Google:**
146+
- `gemini-3-pro`
147+
- `gemini-2.5-pro`
148+
- `gemini-2.5-flash`
149+
150+
**Ollama (Self-hosted):**
151+
- `granite4:3b`
152+
- `qwen3:8b`
153+
- `cogito:latest`
154+
- Any model you have pulled locally
155+
156+
**vLLM (Self-hosted):**
157+
- Any model served by your vLLM instance
158+
85159
## Development
86160

87161
```bash
@@ -112,25 +186,6 @@ devops-agent/
112186
└── docs/ # Documentation
113187
```
114188

115-
## Common issues and fixes
116-
- if you see any error like `INFO Error checking if content_hash ed7002b439e9ac845f22357d822bac1444730fbdb6016d3ec9432297b9ec9f73 exists: Unexpected Response: 400 (Bad Request)
117-
Raw response content:
118-
b'{"status":{"error":"Bad request: Index required but not found for \\"content_hash\\" of one of the following types: [keyword]. Help: Create an index for this key or use a
119-
different filter."},"time":2 ...' `
120-
```text
121-
curl --request PUT \
122-
--url https://9df18135-290c-45b3-8158-f73b103dc352.eu-west-2-0.aws.cloud.qdrant.io:6333/collections/devops-memory/index \
123-
--header 'Authorization: Bearer YOUR_API_KEY' \
124-
--header 'Content-Type: application/json' \
125-
--data '{
126-
"field_name": "content_hash",
127-
"field_schema": {
128-
"type": "keyword",
129-
"on_disk": true
130-
}
131-
}'
132-
```
133-
134189
## Contributing
135190

136191
Contributions are welcome! Please feel free to submit a Pull Request.
@@ -142,12 +197,10 @@ Apache2.0 License - see LICENSE file for details
142197
## RoadMap
143198

144199
- [ ] Implement log analysis with pattern detection
145-
- [ ] Add support for multiple LLM providers with model params `--model gpt-5-mini`
146-
- [ ] Add Support for self-hosted models `Ollama` & `vLLM`
147-
- [ ] Add Support for Reasoning controls
148200
- [ ] Add Support for MCP to use local file system for quick access
149201
- [ ] Add support for Human-in-the-Loop for more focused and collaborated work
150-
- [ ] Create direct pip package for easy install of the agent.
202+
- [ ] Support for custom prompt templates
203+
- [ ] Agent as a Service with privacy first concept
151204

152205
## Support
153206

‎release_notes.md‎

Lines changed: 18 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,18 @@
1+
# DevOps Agent
2+
3+
## What's New in Latest Release 🎉 (31-12-2025)
4+
5+
### New Features
6+
- ✨ **Multiple LLM Provider Support**: Choose between OpenAI, Anthropic, Google Gemini, Ollama, or vLLM
7+
- 🎯 **Dynamic Model Selection**: Specify any model name compatible with your chosen provider
8+
- 🧠 **Reasoning Mode**: Enable advanced reasoning capabilities with `--reasoning_enabled`
9+
- 🐛 **Debug Mode**: Detailed logging for troubleshooting with `--debug_mode`
10+
- 🔒 **Self-Hosted Options**: Full support for Ollama and vLLM for private deployments
11+
- 📦 **PyPI Package**: Now available for direct installation via `pip install devops-agent`
12+
13+
### Improvements
14+
- 🔧 Better error handling and user feedback
15+
- ⚡ More intuitive CLI interface
16+
- 📚 Enhanced documentation and examples
17+
- 🎨 Improved output formatting
18+
- 🧠 Improved Memory for agents

0 commit comments

Comments
 (0)