Understand a large TypeScript codebase — fast

Map the repository in seconds

Paste the GitHub URL into the OpenVisio web app and the repository is fetched and indexed entirely in your browser — nothing uploaded, no account for your first map. For private work, the CLI (npm install -g openvisio) indexes any local folder the same way.

Indexing is deterministic — tree-sitter parses the repository into files, symbols, imports, and calls, and PageRank ranks what matters. No LLM ever sees your code during indexing, nothing is uploaded, and the whole thing runs on your machine.

What the map shows in a TypeScript project

The graph resolves the real import edges behind the aliases and barrels, so you see which packages actually depend on which, where the circular imports hide, and which utility files the whole workspace leans on.

Four views on one graph: a 3D city (folders are districts, files are buildings sized by lines of code and colored by hotness), an atlas constellation of symbols, and architecture and folder dependency views. Filter by language, folder, or recently changed.

Ask, and get cited answers

The narrator explains the architecture and traces any flow in plain words, and every claim carries a path:line citation — touch it and the file lights up on the map. It is grounded in the deterministic graph, not free-form generation.

Give your AI agent the same map

The graph you explore is the same one served to coding agents over MCP — Claude Code, Cursor, Codex. Once connected, the agent has six structured queries instead of blind file reads: resolve_context (a task-ranked skeleton plus the relevant neighborhoods, every line with a path:line anchor), find_symbol, get_dependents, get_neighborhood, get_hotspots, and get_repo_skeleton. Discovery stops being a token sink and becomes a lookup.

Frequently asked questions