Codebase Context via cxpak

When this skill is triggered, gather structured codebase context using cxpak before answering the user's question.

Steps

1. Ask for token budget. Ask the user: "How large a context budget should I use? (default: 50k tokens)" If they don't specify, use 50k.

2. Resolve the cxpak binary. Run the ensure-cxpak script to get the binary path:

   CXPAK="$("${CLAUDE_PLUGIN_ROOT}/lib/ensure-cxpak")"

3. Run cxpak overview. Execute:

   "$CXPAK" overview --tokens <budget> --format markdown .

   Where <budget> is what the user specified (e.g., 50k, 100k, 20k).

4. Use the output. The command outputs a structured codebase summary including:

   • Project metadata (file counts, languages, tokens)
   • Directory tree
   • Module/component map with public symbols
   • Dependency graph (import relationships)
   • Key files (README, configs, manifests)
   • Function/type signatures
   • Git context (recent commits, churn)

5. Answer the user's question using the cxpak output as your primary source of codebase understanding.

6. Mention the source. Tell the user that cxpak provided the structured codebase context, e.g., "Based on the cxpak overview of this repo..."

Important

• Always use --format markdown — it's native to your context window.
• If cxpak fails (not a git repo, no files found), fall back to standard file reading.
• The overview may include .cxpak/ detail file pointers if the repo exceeds the budget — you can read those for deeper context.