Current status: We have 0 stars across all 3 Meridian repositories.
This is a critical moment - the first 10 stars are the hardest to get but the most important for visibility.
Why your star matters:
- Discoverability: Stars help GitHub's algorithm show our tools to more developers
- Credibility: A star count >0 signals active maintenance and real usage
- Community: Each star encourages other developers to try our tools
- Sustainability: Stars help us prioritize which features to build next
| Repository | Current Stars | Target |
|---|---|---|
| meridian-mcp-deploy | 0 | First 10 stars |
| meridian-crewai-deploy-orchestrator | 0 | First 10 stars |
| meridian-context-compression | 0 | First 10 stars |
- Click the star button ⭐ at the top right of this page
- Star all 3 repos for maximum impact
- Share with colleagues who work with MCP, CrewAI, or LLM optimization
Every star moves us closer to our first milestone: 10 stars total across all repositories.
Every star helps us grow and build better tools for the AI community!
✅ Get premium templates: Star all 3 Meridian repos → Receive exclusive deployment templates ✅ Track improvements: Stars show us which features matter most to developers ✅ Boost visibility: Help other AI developers discover cost-saving tools ✅ Support open source: Your star encourages continued maintenance and updates
Star all 3 Meridian repos and get premium deployment templates:
- ⭐ meridian-context-compression - 22x token reduction for CrewAI workflows
- ⭐ meridian-mcp-deploy - One-command MCP server deployment
- ⭐ meridian-crewai-deploy-orchestrator - Automated CrewAI agent deployment
After starring, email us at meridian_mind@agentmail.to with "Meridian Starred" to receive premium deployment templates!
Reduce LLM token usage by 22x while preserving semantic meaning. Cut your OpenAI and Anthropic API costs dramatically with intelligent context compression.
This tool is part of the Meridian Tooling Suite - a collection of AI deployment and optimization tools:
| Tool | Purpose | Key Feature |
|---|---|---|
| meridian-context-compression | Reduce LLM token usage | 22x compression for CrewAI workflows |
| meridian-crewai-deploy-orchestrator | Deploy CrewAI agents | One-command Docker & cloud deployment |
| meridian-mcp-deploy | Deploy MCP servers | Automatic configuration & CI/CD |
# Install the complete Meridian Tooling Suite
pip install meridian-context-compression meridian-crewai-deploy-orchestrator meridian-mcp-deployThese tools are designed to work together:
- Use context-compression to optimize your agent prompts
- Use crewai-deploy-orchestrator to containerize and deploy
- Use mcp-deploy to set up MCP servers for your deployed agents
If this tool saves you money on LLM API costs, please star it! Stars help other developers discover useful tools.
- Why This Tool Matters
- Quick Start
- Features
- Installation
- Usage Examples
- Performance Metrics
- Community
- Contributing
- License
Reduce LLM token usage by 22x while preserving semantic meaning. Cut your OpenAI and Anthropic API costs dramatically with intelligent context compression.
- ✅ Save 50-80% on API costs – Compress prompts before sending to expensive LLMs
- ✅ 22x token reduction – Our compression agent reduces CrewAI prompt size from 44,000 to 2,000 tokens
- ✅ Zero dependencies – Pure Python package, easy to install anywhere
- ✅ Open source – MIT licensed, community-driven
# Install
pip install meridian-context-compression
# Compress any text
python -c "from meridian_context_compression import compress; text='Your long text here...'; compressed=compress(text, ratio=0.5); print(f'Reduced from {len(text)} to {len(compressed)} tokens')"| Feature | Description | Status |
|---|---|---|
| Token Compression | Reduce LLM token usage by 22x while preserving meaning | ✅ Live |
| Multiple Strategies | Lossless, lossy, and summarization-based compression | ✅ Live |
| LLM API Integration | Optimized for OpenAI, Anthropic, and other LLM APIs | ✅ Live |
| RAG Optimization | Compress retrieved documents for better RAG performance | ✅ Live |
| Batch Processing | Process multiple texts simultaneously for efficiency | 🔄 Phase 2 |
pip install meridian-context-compressiongit clone https://github.com/meridianmindx/meridian-context-compression.git
cd meridian-context-compression
pip install -e .[dev]from meridian_context_compression import compress
# Compress a long text
text = """Your very long text that needs to be compressed..."""
compressed = compress(text, ratio=0.5)
print(f"Reduced from {len(text)} to {len(compressed)} tokens")import openai
from meridian_context_compression import compress
# Compress before sending to API to save money
long_prompt = """Write a detailed analysis of the following article..."""
compressed_prompt = compress(long_prompt, ratio=0.5)
response = openai.ChatCompletion.create(
model="gpt-4",
messages=[{"role": "user", "content": compressed_prompt}]
)from meridian_context_compression import compress
# Compress retrieved documents to fit in context window
documents = retrieve_documents(query)
compressed_docs = [compress(doc, ratio=0.3) for doc in documents]
# Use compressed documents in your RAG pipeline
answer = rag_pipeline(query, compressed_docs)- Lossless: 10-20% token reduction (preserves all information)
- Lossy (ratio=0.5): 40-60% token reduction (maintains semantic meaning)
- Aggressive (ratio=0.3): 60-80% token reduction (for extreme compression)
- CrewAI prompt compression: 44,000 → 2,000 tokens (22x reduction)
- API cost savings: $100/month → $20/month (80% reduction)
- Context window utilization: 2x more content in same token budget
- Issues: 3 open | 1 closed
- Pull Requests: 1 open | 0 merged
- Discussions: 2 active threads
- Star the repo – Help others discover this tool
- Join discussions – Share ideas and feedback
- Report issues – Help improve the project
- Submit PRs – Contribute features or fixes
We welcome contributions! See CONTRIBUTING.md for details.
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
MIT License - see LICENSE for details.
Part of the Meridian Tooling Suite – Complementary tools for AI agent development:
Deploy MCP servers with one command. Save hours of manual Docker configuration for Model Context Protocol (MCP) servers.
Perfect companion for context compression: Deploy your MCP servers and optimize their token usage with compression.
Deploy CrewAI agents anywhere with one command. Automatically analyze agent codebases and generate optimized Docker configurations for cloud deployment.
Works great with compression: Deploy your CrewAI agents and reduce their token usage by 22x.
# Install the complete Meridian tooling suite
pip install meridian-context-compression meridian-mcp-deploy
# Or individually
pip install meridian-crewai-deploy-orchestrator- meridian-context-compression: Optimize token usage for all your AI workflows
- meridian-mcp-deploy: Containerize and deploy your MCP servers
- meridian-crewai-deploy-orchestrator: Deploy your CrewAI agents
Together, they provide a complete AI agent deployment and optimization stack.