onboard

name	onboard
description	First-time Loom setup for a new repository. Runs loom analyze, installs MCP config, and shows what was indexed.
argument-hint	[repo-path]
allowed-tools	["mcp__loom__graph_stats","mcp__loom__god_nodes","mcp__loom__suggest_questions","mcp__loom__get_savings","Bash","Read","Glob"]

Onboard — First-Time Loom Setup

Set up Loom for a repository and orient to the codebase.

Steps

1. Check if already indexed

Call graph_stats() — if total node count > 0, Loom is already indexed. Jump to Step 4.

2. Index the repository

loom analyze .

This runs tree-sitter over all Python/TypeScript/JavaScript/Java files and builds ~/.loom/projects/{repo-name}.db (or ~/.loom/loom.db if not in a git repo). Typically 5–30 seconds for a 500-file repo.

If the user provides a path argument, use that instead of ..

3. Install MCP config (first time only)

loom install

This writes the MCP server config to the appropriate tool config file (Claude Code, Cursor, Windsurf, or Codex), and installs a post-commit git hook for incremental sync.

4. Check what was indexed

Call graph_stats() — shows node and edge counts by kind.

5. Find entry points

Call god_nodes(limit=10) — shows the most-called functions. These are the best starting points for understanding the codebase.

6. Surface investigation priorities

Call suggest_questions(limit=7) — surfaces dead code, undocumented hot functions, bridge nodes, and low-cohesion clusters worth looking at.

7. Report to user

Show:

• Node count by kind (functions, classes, files, methods)
• Top 5 most-called functions (entry points)
• Top 3 suggested questions to investigate
• Confirm MCP server is registered (loom in mcpServers)