Project Nodal — AI skill for Claude Code
A local-first, infinite canvas that turns linear AI chats into a spatial knowledge graph.
How to install Project Nodal
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open yibie/project-nodal and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Project Nodal does
A local-first, infinite canvas that turns linear AI chats into a spatial knowledge graph.
Alternatives in AI
- Wiki Graph — Build the LLM Wiki knowledge graph 2.3k ★
- Nodeterm — Node-based terminal manager for AI coding agents — tmux-backed terminals and parallel agent sessions as dragga 1.3k ★
- Memory Forge Rs — Stop resetting satisfying AI chats — edit the memory instead 474 ★
README
Project Nodal: Chat is a Line, Thought is a Network.
**🎥 [Watch the 15s Demo Video (MP4)](image/demo.mp4)**
**The "Chat Interface" is the biggest lie in modern computing.**
Open ChatGPT, Claude, or Gemini. What do we see? A linear feed. A text stream. A modernized teletype machine from the 1970s.
They force our brains into a single-threaded, chronological queue. But **that is not how we think.**
Our minds are graphs. They jump from a coding bug, to a deployment strategy, to a marketing slogan, and back to the bug. It is concurrent, messy, and spatial. Forcing networked brains into a linear interface is cognitive suffocation.
We built **Project Nodal** to break the timeline.
1. Space > Time
In standard AI, "Context" is just the previous 50 messages. If we change the subject, we pollute the context. In **Project Nodal**, context is **Spatial**.
- The Infinite Canvas: Drag the viewport. Zoom out to see the big picture. Zoom in to focus on a detail.
- Zones: Create a "Zone" for backend logic. Create another for frontend design. Move them apart. Visually group our thoughts.
2. Objects > Streams
In Nodal, a conversation isn't a fleeting moment in a scroll; it's a **Persistent Object** (A Sticky Note).
- Multi-Threaded: Have 5 conversations with the AI simultaneously.
- Drag & Drop: Treat thoughts like physical objects. Move them. Resize them. Collapse them when they are "done" but keep them visible.
3. Fork Thoughts
This is our "Killer Feature". In linear chat, if we want to explore a "What If" scenario, we often have to derail the whole conversation. In **Nodal**, just click **Fork**.
- Branching: Take a specific message, spin it off into a new Note, and explore that path without losing the original context.
- Exploration: Go down the rabbit hole. If it's a dead end, just delete the note. The original conversation remains pure.
4. Synapses > Lists
- Connections: Draw a line from a "Product Idea" note t
Related Skills
Graphmind
Local-first code intelligence for AI assistants. Turns your codebase into a knowledge graph your AI can query,
Quilt
AI-native infinite-canvas design tool: describe an app, get multi-screen HTML designs on an infinite canvas, c
Graph Hub
Local-first code intelligence for AI agents. Transform your codebase into a queryable knowledge graph. 85% tok
Mirrordraw Devkit
MirrorDraw developer kit: MCP server, plugin SDK and scaffolding CLI to let AI agents (Claude, Cursor, Codex)
Learnlance
Learn from your AI coding agents. Learnlance turns AI-generated code into a personal knowledge graph of concep
OpenSRE
OpenSRE — the memory-first AI SRE. Self-hosted incident investigation with episodic memory, knowledge graph, w
Related Agents
49agents
49Agents bills itself as the first 2D agentic IDE — all agents, all terminals, all projects, all machines, one
Explainer Director
Delegate to this agent to make a vertical first-person making-of video (3-5 min, 公众号/视频号/抖音) with real-data ca
Fire Learncoding Walker
Dependency graph mapper for learncoding mode — detects entry point and produces ordered linear step list from