简体中文 | English
This repository contains Siada CLI, a command-line AI workflow tool that provides specialized intelligent agents for code development, debugging, and automation tasks.
Siada CLI is context-aware of your project, computing environment, and cross-session persistent memory, offering consistent, personalized, and continuously improving intelligent assistance across Terminal, Feishu (Lark), and Web interfaces.
Current latest version: v1.7.0
- Code Generation — Create new features, web interfaces, and refactor code with multi-language support
- Bug Fixing — Automatically identify, analyze, and fix code defects in large codebases
- Long-Horizon Task Execution — Tackle complex, multi-step development tasks through spec-driven execution and sub-agent orchestration, keeping goals on track across long task chains
- Self-Evolution — Persistent memory and scheduled proactive tasks keep the assistant continuously improving
- Host Management — Remote management and processing of your local machine and documents
- Multi-Entry Support — Use in Terminal (interactive/non-interactive), Feishu remote control, or Web UI
| Requirement | Details |
|---|---|
| OS | macOS, Linux |
| GCC | 11+ |
| Package Manager | uv |
Step 1: Install uv
curl -LsSf https://astral.sh/uv/install.sh | shStep 2: Install Siada CLI
uv tool install --force --python python3.12 --compile --with pip siada-cli@latestIf siada-cli is not found on PATH after installation:
uv tool update-shelluv tool upgrade siada-cliuv tool uninstall siada-cli-
Prerequisites: Ensure you have Python 3.12+ and Poetry installed.
-
Clone and Install:
git clone https://github.com/liauto-siada/siada-cli.git cd siada-cli poetry install -
Build UI (required before first run):
cd siada_cli_ui && npm ci && npm run build:all && cd ..
-
Run CLI:
# Method 1: Run with Poetry poetry run siada-cli # Method 2: Activate virtual environment (recommended) source $(poetry env info --path)/bin/activate siada-cli
On first launch, Siada CLI will guide you through an interactive setup — simply select your provider, choose a model, and enter your API key. No manual configuration needed to get started.
For advanced configuration (config file, environment variables, CLI flags), see the User Manual.
For custom external models (private deployments), see External Model Configuration.
siada-clisiada-cli --no-interactive "Fix the null pointer exception in src/main.py"Connect Siada CLI to your Feishu (Lark) workspace and send tasks remotely. See Remote Control Guide.
Cross-session persistent memory lets Siada remember your preferences, codebase context, and past decisions — making every interaction more accurate over time.
Save and restore agent states at key decision points, giving you full control over long-running tasks.
Define reusable instruction sets and workflows as SKILL.md files that extend Siada's capabilities for your specific domain or team conventions. See Skills Documentation.
Connect to external tool servers via the Model Context Protocol (MCP) to extend Siada's toolset.
Run Siada as a background daemon to execute proactive, scheduled automation tasks.
Place siada_rule.md or .siadarules/ files in your project or home directory to set persistent AI behavior rules without repeating yourself in every conversation.
Intelligently compress conversation history to stay within model context limits without losing critical context.
| Document | Description |
|---|---|
| User Manual | Full usage guide, commands, and configuration |
| Skills System | Create and manage reusable AI skills |
| External Model Config | Configure private/custom model endpoints |
| Remote Control (Feishu) | Control Siada via Feishu messages |
| Compaction Configuration | Context compression strategy settings |
| Contributing Guide | How to contribute to this project |
We welcome contributions to Siada CLI! Whether you want to fix bugs, add new features, improve documentation, or suggest enhancements, your contributions are greatly appreciated.
Please read our Contributing Guide before submitting:
- Project vision and development goals
- Directory structure and development guidelines
- Pull request guidelines and best practices
- Code organization principles
Siada CLI is built upon the foundation of numerous excellent open source projects. We extend our deepest respect and gratitude to their contributors.
Special thanks to:
- OpenAI Agents SDK — foundational agent framework
- LiteLLM — unified LLM provider interface
- OpenHands ACI — agent-computer interface tooling
- Gemini CLI — inspiration for terminal-native AI agent design
For a complete list of open source projects and licenses, see CREDITS.md.
Distributed under the Apache 2.0 License.
See disclaimers.md.
