TORCH
A Simple Round-Robin Agent Scheduler.
A lightweight, drop-in folder that rotates through autonomous agents to maintain and improve your codebase. No complex pipelines—just simple, effective orchestration.
Choose Your Agent
Web-based tool with Auto-Scheduler. Optimized for Google's latest coding agent. Learn more.
Advanced reasoning for complex tasks. (Requires full access permissions) Learn more.
Open source models running locally. Learn more.
Long-context understanding and safety. (CLI Required, needs cronjob) Learn more.
Google's multimodal AI model. (CLI Required, needs cronjob) Learn more.
Alibaba's open weights coding model. (CLI Required, needs cronjob) Learn more.
The autonomous Python IDE. (Manual only) Learn more.
Copy this happy-path prompt to onboard TORCH:
Requires Node.js 22+.
Hey Jules, can you install and setup TORCH into this repo? Here is the install command:
npm install https://github.com/PR0M3TH3AN/TORCH/archive/refs/heads/main.tar.gz && npx --no-install torch-lock init
Running Offline or Air-Gapped?
Download the standalone bundle and install it directly without registry access.
Download Bundlenpm install ./{{OFFLINE_BUNDLE_FILENAME}}
Why Run TORCH?
Drop-in Orchestration
Transform any LLM session into a productive workforce. Just paste a meta-prompt, and watch it coordinate tasks autonomously.
Verifiable Handoffs
Code doesn't ship unless it works. Agents must pass your specific tests and linters before releasing the lock.
Decentralized State
Serverless coordination via Nostr. No databases to manage, just pure, distributed task locking across agents.
Persistent Memory
Agents that remember. TORCH maintains a semantic history of decisions, preventing loops and regressions.
Hook memory into your existing agents
Add these instructions to your agent's system prompt (e.g. in AGENTS.md or CLAUDE.md) to enable the TORCH memory system.
## TORCH Memory Integration
You have access to the TORCH memory system.
1. READ: Check `.scheduler-memory/latest/${cadence}/memories.md` for past learnings.
2. WRITE: Before exiting, save new insights to `memory-update.md` so future runs can learn from this session.
Functional Agents
Loading agents...