scorp2006

Nodes — AI skill for Claude Code

AI community

Run AI coding agents in parallel without them breaking each other’s work.

How to install Nodes

This entry records only its repository, not the path inside it, so there is no exact command to give. Open scorp2006/Nodes and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Nodes does

Run AI coding agents in parallel without them breaking each other’s work. A shared FalkorDB graph gives every agent atomic ownership of the code it edits, checks facts against their sources, and catches deadlocks, all over MCP and Claude Code hooks, with a live dashboard.

Alternatives in AI

Nodeterm — Node-based terminal manager for AI coding agents — tmux-backed terminals and parallel agent sessions as dragga 1.3k ★
Memmy Agent — 🍙 A personal AI agent & local memory hub for all AI agents, gives every AI one shared, fully controlled memor 1k ★
Cwc Long Running Agents — Claude Code's built-in /goal command gives you a generator/evaluator loop out of the box: set a completion con 668 ★

README

Nodes

**Run your AI agents in parallel without them breaking each other's work.**

[![License: MIT](https://img.shields.io/badge/license-MIT-blue)](LICENSE) [![CI](https://github.com/scorp2006/Nodes/actions/workflows/ci.yml/badge.svg)](https://github.com/scorp2006/Nodes/actions/workflows/ci.yml) [![FalkorDB 4.22.1](https://img.shields.io/badge/FalkorDB-4.22.1-e04a3a)](https://www.falkordb.com/) [![MCP](https://img.shields.io/badge/MCP-server-2f7fd1)](https://modelcontextprotocol.io/)

[Demo video](DEMO_VIDEO_URL) · [Blog post](https://dev.to/scorp2006/from-12-to-38-how-a-graph-turned-claude-subagents-into-a-team-1h3n) · [Setup](SETUP.md)

![The Nodes dashboard: three agents, two hand-offs, six conflicts prevented, and a plain-English feed of what happened](docs/img/dashboard.png)

The problem

Give several coding agents one repo and they collide: same function, stale assumptions, unmergeable work. Anthropic saw it in its own agent swarms: *"The PRs often conflicted with one-another, at which point they were then abandoned"* ([Patterns and problems in emerging multiagent systems](https://www.anthropic.com/research/multiagent-systems)). Across 107k agent pull requests on GitHub, **27.67%** hit merge conflicts ([AgenticFlict](https://arxiv.org/abs/2604.03551)).

Results

**Pre-registered, held-out, same model in every condition.**

**HiddenBench**, 40 held-out tasks, Claude Haiku 5.5: groups decide correctly **38%** of the time with Nodes, vs 15% chat, 12% placebo, 12% board without source checks. Primary test vs placebo **p = 0.006** ([result](results/test1_primary.txt), [pre-registration](results/preregistration_test1.md)).

**CooperBench**, 45 feature pairs, Claude Sonnet 5.5: two agents' work merges cleanly **35/45** with Nodes vs 22/45 with the same procedure over chat, Holm **p = 0.002** ([result](results/cb1b_primary.txt), [pre-registration](results/preregistration_cb1b.md)).

**Every proof in one place:** [results/README.md](results/REA