🥷 AI agents for knowledge work, not code
This repository also configures its own agents. See what kage-bunshin tells them →
Agent Claude Code
Analyst agent. Handles critical analysis, risk assessment, quality reviews, and feasibility evaluation. Use when asked "what do you think?", "any issues?", "review this", "evaluate this", "give feedback", "SWOT analysis", "what are the risks?", or "check this".
Agent Claude Code
Ideator agent. Handles creative brainstorming, divergent thinking, and new idea generation. Use when asked for ideas, "any interesting suggestions?", new feature proposals, solution brainstorming, or creative suggestions.
Agent Claude Code
Planner agent. Handles task breakdown, action plan creation, milestone definition, and dependency mapping. Use when asked to plan, create schedules, break down tasks, create roadmaps, or "how should I proceed?" and "make a plan".
Agent Claude Code
Researcher agent. Handles deep investigation, information gathering, source triangulation, and synthesis. Use when asked to research, investigate, do technical/market research, comparative analysis, or "look into..." and "what's the current state of...?".
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: