Skip to content

mirasoth/soothe

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

1,308 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

✨ Soothe is an agent-harnessing frameworkβ€”an Agentic OS that pushes humans out of the execution loop. Built on LangChain / DeepAgents, it adds a persistent agentic loop and goal engine that maintains context across sessions, sustains long-running goals, coordinates multiple objectives, and autonomously steers complex tasks. Shift from human-in-the-loop to agent-in-the-loop: define intent, let the system handle execution.


πŸš€ Key Features

  • ✨ Thinks Ahead β€” Multi-step planning with dynamic adaptation
  • πŸš€ Acts Autonomously β€” Research, coding, file ops, plugin automation
  • 🧠 Learns & Remembers β€” Persistent memory across sessions
  • πŸ”’ Stays Secure β€” Least-privilege, local-first architecture
  • πŸ”Œ Extends Easily β€” Decorator-based plugins, custom tools, subagents
  • 🌐 Works Anywhere β€” Multi-transport daemon (WebSocket, HTTP REST)

Design Philosophy

Principle Description
Plan β†’ Execute Autonomous loop: plan, act, evaluate, adapt
Persistent Memory Resume threads, recall context, track goals
Security First Local execution, least-privilege policies
Plugin Architecture Decorator-based tools, subagents, MCP servers

What Can Soothe Do?

Capability Features
Deep Research Multi-source web search, academic papers (arXiv, DeepXiv), document analysis
Autonomous Execution Multi-step workflows, file ops, code execution, shell commands
Long-Running Ops Background daemon, thread management, persistent state
Custom Plugins @tool, @subagent, @plugin decorators, MCP server integration

Milestones

Status Milestone
βœ… Single-Session Autonomy β€” End-to-end goal execution
βœ… Cross-Thread Continuity β€” Persistent context across threads
βœ… Multi-Goal Orchestration β€” Interdependent long-horizon workflows
⏳ Benchmark Reproduction β€” Compiler experiment

Getting Started

Choose one of these two ways to run Soothe:

1) Recommended: Docker deployment (deploy/)

Fastest path for daily use.

cd deploy
# Configure env + keys (see deploy/README.md)
docker compose up -d

Then use:

soothe
# or
soothe -p "Research top 5 Python web frameworks"

2) Alternative: Install from pip

Requirements: Python 3.11+.

pip install -U 'soothe[all]' soothe-cli soothe-daemon
cp config/config.template.yml ~/.soothe/config/config.yml
export OPENAI_API_KEY="sk-..."  # or ANTHROPIC_API_KEY / DASHSCOPE_API_KEY
soothed start
soothe

Documentation

Resource Description
User Guide End-user usage guide
RFCs Architecture specs
CLAUDE.md AI agent dev guide

License

MIT

About

πŸ›€ Soothe - An autonomous expert system engineered to orchestrate complex workflows and execute long-duration tasks with precision and persistence.

Topics

Resources

License

Stars

Watchers

Forks

Packages

 
 
 

Contributors

Languages