juanje

Generate AI reviewer context files that prevent outdated dependency suggestions, internal library confusion, and architectural misunderstandings. Use when setting up AI code review (ai-code-review, Cursor, etc.), refreshing stale context, or creating team coding standards. Handles .ai_review/project.md generation and updates.

Build and maintain a local, portable, markdown-based knowledge base. Use when the user wants to ingest documents into a wiki, check wiki health, synthesize emergent concepts, or file ephemeral knowledge. Trigger on: "ingest", "add to knowledge base", "add to wiki", "build wiki", "process documents into wiki", "lint wiki", "check wiki", "wiki health", "review the wiki", "synthesize", "find patterns", "what concepts are missing", "file this to wiki", "capture this to wiki", "import into wiki", "ingest this URL", "organize wiki", "fix wiki tags"

Use this skill when the user wants to understand, document, or explain a codebase — even if they don't say "document" explicitly. Trigger on: "explain this codebase", "how does this project work", "create onboarding docs", "document the architecture", "I'm new to this repo", "generate a codebase overview", or any request to understand unfamiliar code. Produces a single docs/CODEBASE.md with architecture, components, data flow, setup instructions, and a recommended reading path for new developers. Supports quick/standard/deep depth levels and incremental updates.