Terminal Injected Coding CLI Assistant
Project description
Ticca
Terminal Injected Coding CLI Assistant
A privacy-focused AI coding assistant for your terminal. Multi-model support, specialized agents, and zero telemetry.
Quick Start
# Using UV (recommended)
curl -LsSf https://astral.sh/uv/install.sh | sh
uvx ticca
# Using pip
pip install ticca
ticca
Features
- 🎨 Beautiful TUI - Terminal interface powered by Textual
- 🤖 Multi-Model - OpenAI, Claude, Gemini, Cerebras, Ollama, custom endpoints
- 🎯 Specialized Agents - Code review, debugging, security auditing, refactoring
- 🔌 MCP Integration - Extend with Model Context Protocol servers
- 🔒 Privacy First - Zero telemetry, local-only option, no data collection
- ⚡ Load Balancing - Round-robin across multiple API keys
Usage
ticca # Start TUI
ticca -i # Interactive CLI
ticca -p "Explain this code" # Single prompt
ticca -m gpt-4 -a code-reviewer # Specify model and agent
In-Session Commands
/agent <name> # Switch agent (code-reviewer, security-auditor, etc.)
/model <name> # Switch model
/mcp list # Manage MCP servers
/set <key> <value> # Configure settings
/help # Show help
Configuration
API Keys
Set via environment variables or TUI settings:
export OPENAI_API_KEY=sk-...
export ANTHROPIC_API_KEY=sk-ant-...
export GEMINI_API_KEY=...
Models
Configure in ~/.ticca/models.json or ~/.ticca/extra_models.json:
{
"gpt-4": {
"type": "openai",
"model": "gpt-4",
"max_tokens": 8000
}
}
Custom Commands
Create markdown files in .claude/commands/, .github/prompts/, or .agents/commands/:
cat > .claude/commands/review.md << 'EOF'
# Code Review
Review for security, performance, style, and best practices.
EOF
/review # Use in Ticca
Agents
Built-in specialized agents:
- code-reviewer - Code quality and best practices
- security-auditor - Security vulnerability analysis
- debugger - Bug identification and fixes
- refactorer - Code improvements
- documenter - Documentation generation
Create custom agents in ~/.ticca/agents/ as JSON files or use /agent agent-creator.
MCP Servers
Extend capabilities with Model Context Protocol:
/mcp list # Show servers
/mcp start <server> # Start server
/mcp status # Check status
Configure in ~/.ticca/mcp_servers.json:
{
"file-server": {
"enabled": true,
"path": "/usr/local/bin/mcp-file-server",
"config": {}
}
}
Advanced Features
Load Balancing
Rotate across multiple API keys:
{
"gpt4-balanced": {
"type": "round_robin",
"models": ["openai-key-1", "openai-key-2"],
"rotate_every": 5
}
}
Custom Endpoints
Self-hosted or private models:
{
"local-llama": {
"type": "openai",
"model": "llama-2",
"custom_endpoint": {
"url": "http://localhost:8000/v1",
"api_key": "not-needed"
}
}
}
DBOS (Optional)
Durable execution with automatic recovery:
/set enable_dbos true
Privacy
✅ Zero telemetry or tracking
✅ No cloud storage of conversations
✅ Local-only mode available (Ollama)
✅ Direct API communication (no proxies)
✅ Your code never leaves your machine
Requirements
- Python 3.11+
- One of: OpenAI, Anthropic, Gemini, Cerebras API key, or local LLM (Ollama, VLLM)
Installation
UV (Recommended)
curl -LsSf https://astral.sh/uv/install.sh | sh
export UV_MANAGED_PYTHON=1
uvx ticca
pip
pip install ticca
ticca
From Source
git clone https://github.com/janfeddersen-wq/ticca.git
cd ticca
./start.sh
Contributing
Contributions welcome! Please:
- Follow existing code style
- Include tests (
pytest) - Update documentation
- Maintain backward compatibility
See CONTRIBUTING.md for details.
License
MIT License - see LICENSE for details.
Links
- Repository: https://github.com/janfeddersen-wq/ticca
- Issues: https://github.com/janfeddersen-wq/ticca/issues
- Discussions: https://github.com/janfeddersen-wq/ticca/discussions
- AGENT.md: https://agent.md
- MCP: https://modelcontextprotocol.io
Acknowledgments
Ticca builds upon the excellent work of:
- Code Puppy - The awesome foundation that Ticca is built upon
- gac - The awesome git message generator (soon to be integrated)
Ticca - Built with ❤️ for developers who value efficiency, privacy, and simplicity.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file ticca-0.1.4.tar.gz.
File metadata
- Download URL: ticca-0.1.4.tar.gz
- Upload date:
- Size: 323.2 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.9.5
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
010084c10529db91099699d116819099b5a0f951191f972dadd108f8555e937f
|
|
| MD5 |
0d727fbb8c6add90cff8a91db0159d2c
|
|
| BLAKE2b-256 |
37dabc5a7512fe4d87659dcf1acd9956077962f080ffc5fa35b91214ef48cccc
|
File details
Details for the file ticca-0.1.4-py3-none-any.whl.
File metadata
- Download URL: ticca-0.1.4-py3-none-any.whl
- Upload date:
- Size: 420.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.9.5
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
0faaba151991e0ff1090440b9b512d9b7cedebd296f72ad072ce01f0e36f4bad
|
|
| MD5 |
486f42de2444f7c51f9d5f71c3524e7a
|
|
| BLAKE2b-256 |
4b9b6a62b9ba99f0c957fa231c27e6af853a3e9612ee69e0b42a01362b2ec9d8
|