Execute six-stage causal reasoning from Input to Learning for operational decisions.
npx skills add https://github.com/BellaBe/lean-os --skill reasoning-causal
Please help me install this Agent Skill. Skill: reasoning-causal Source: https://github.com/BellaBe/lean-os/tree/main/.claude/skills/reasoning-causal Command: npx skills add https://github.com/BellaBe/lean-os --skill reasoning-causal
Executes structured cause-and-effect reasoning to inform decisions and actions.
Provide a Stage 1 observation; returns initial hypothesis and plan.
High-intent search queries and answers about installing and using this skill.
Causal reasoning applies a 6-stage flow—Input, Hypothesis, Implication, Decision, Actions, Learning—to systematically trace cause-effect chains in known processes. Each stage produces artifacts and explicit reasoning steps, turning observations into evidence-based decisions and updating your goals with outcomes.
Canvas hypothesis testing lets you structure assumptions upfront before deciding; reactive threads respond to immediate observations. Both approaches feed the same 6-stage flow, so you choose based on whether you're testing planned scenarios or responding to live operational changes.
Yes. Thread binding connects your reasoning directly to goal or subgoal threads, so outcomes automatically update your goals and track learning across decisions. This keeps cause-effect analysis tied to what you're actually trying to achieve.
Provide a Stage 1 observation—any operational fact or trigger. The Skill returns an initial hypothesis and action plan, then guides you through the remaining five stages with stage-specific prompts and output formats.
Use causal reasoning when decisions involve explicit cause-effect chains in known processes and you need to track reasoning, test assumptions, and learn from outcomes. It's designed for operational execution where evidence and stage-by-stage artifacts matter more than speed.
BellaBeDecompose user objectives into subgoals, milestones, and success criteria.
yohayetsionStructure outcome reviews comparing outputs to goals and capturing learnings.
jsagirTrack grant opportunities across discovery, research, application, and submission stages.
panaversityImplement structured perceive-decide-act-verify loops with state management and termination logic.
