GitHub user vladignatyev released brain-map-skill, a Python‑only tool that converts an Obsidian vault or gbrain export into a self‑contained interactive HTML map. The output shows a force‑directed graph coloured by top‑level folder, a scrub‑able timeline of note creation, and a detail panel for each node. A prebuilt demo with 992 fictional notes ships ready to open in any browser and supports multiple AI agents.

The builder reads plain Markdown, using YAML front‑matter for tags and timestamps and resolving [[wikilinks]] into edges. Optional dependencies (networkx, numpy, scipy) speed up layout calculation, but the script runs on the standard library alone. Layouts pre‑computed with Cytoscape’s cose algorithm render instantly, even for thousand‑node graphs. This makes it suitable for personal knowledge graphs and team documentation pipelines.

Users can launch the demo with a single file open command on macOS, Linux or Windows, then scrub the timeline to watch the knowledge base grow, filter by theme or type, and search for keywords. The skill can be dropped into Claude Code, OpenAI Codex, Cursor or Gemini agents by cloning the repo into the skills directory. Server or embeddings are required, the workflow simple.