Skillget

CLIver

mcp
Guvenlik Denetimi
Uyari
Health Uyari
  • License — License: Apache-2.0
  • Description — Repository has a description
  • Active repo — Last push 0 days ago
  • Low visibility — Only 6 GitHub stars
Code Gecti
  • Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Gecti
  • Permissions — No dangerous permissions requested
Purpose
This tool is a general-purpose AI agent designed for the command line. It allows users to interact with various Large Language Models (LLMs) and automate complex multi-step workflows using customizable skills and tool integrations.

Security Assessment
The overall risk is rated as Medium. Because it is an autonomous terminal agent, it inherently processes user prompts, executes shell commands, and makes network requests to external LLM APIs (like OpenAI). While a light code scan found no hardcoded secrets or dangerous patterns, users must supply API keys (e.g., via environment variables). The tool does feature a layered permission system to govern command execution, but caution is still advised when running it in auto-accept or "yolo" modes, as autonomous AI actions can be unpredictable.

Quality Assessment
The project is highly active, with its most recent code push occurring today. It uses the standard Apache-2.0 license and includes robust documentation. However, it currently has very low community visibility with only 6 GitHub stars, meaning it has not been broadly tested or vetted by the open-source community.

Verdict
Use with caution — the code is clean and recently maintained, but its low community adoption and inherent ability to autonomously execute terminal commands require careful oversight.
SUMMARY

General-purpose AI agent for your terminal — safe, controlled, and adaptable to any domain.

README.md

CLIver

CLIver is a general-purpose AI agent for your command line. It is not tied to any specific domain — with customizable system prompts, skills, workflows, and MCP integrations, you can adapt it to any task: coding, DevOps, research, writing, data analysis, or anything else you need.

CLIver is also built to be safe and controlled. A layered permission system governs every tool execution, and a structured workflow engine keeps complex tasks on track — so you get the power of autonomous AI without scattered, unpredictable behavior.

Quick Start

# Install via pip
pip install cliver

# Or run with Docker
docker run --rm -it --user $UID:0 -v ~/.cliver:/home/cliver/.cliver \
  -e OPENAI_API_KEY ghcr.io/cliver-project/cliver

# Start chatting
cliver chat "What time is it in Beijing and London?"

# Interactive mode
cliver chat

Key Features

  • Multi-provider LLM support — OpenAI-compatible (Qwen, DeepSeek, etc.), Ollama, vLLM
  • MCP integration — Connect to any MCP server for extended tool capabilities
  • Skills — LLM-driven skill activation following the Agent Skills specification
  • Memory & Identity — Persistent knowledge and CliverProfile management across sessions
  • Permissions — Layered tool permission system (default, auto-edit, yolo modes)
  • Workflows — LangGraph-powered multi-step workflow engine with subagent isolation and pause/resume support
  • Cost Tracking — Config-based pricing per provider and model (no hardcoded tables)
  • Embeddable — AgentCore API can be used as a Python library

Documentation

Full documentation is available at the docs site:

Development

make init      # Set up dev environment
make test      # Run tests
make lint      # Lint + format check
make format    # Auto-fix lint and formatting

License

See LICENSE for details.

Yorumlar (0)

Sonuc bulunamadi