Enforce structured DOING/EXPECT plans and AFTER reviews for tool calls.
npx skills add https://github.com/acrulopez/dotfiles --skill reasoning-protocol
Please help me install this Agent Skill. Skill: reasoning-protocol Source: https://github.com/acrulopez/dotfiles/tree/main/.claude/skills/reasoning-protocol Command: npx skills add https://github.com/acrulopez/dotfiles --skill reasoning-protocol
Explicit reasoning protocol that enforces structured thinking before and after every action, reducing errors and missteps during tool calls, file operations, code changes, and debugging.
Describe the action to be taken, predict the outcome, specify the next steps for both yes and no cases, then perform the action.
High-intent search queries and answers about installing and using this skill.
Structured thinking before tool calls prevents errors by requiring an explicit plan outlining what you are DOING and what you EXPECT, ensuring actions align with predicted outcomes before execution.
To implement a reasoning protocol for debugging, outline your DOING, EXPECT, IF YES, and IF NO steps before acting, then perform an AFTER review checking RESULT, MATCHES, and THEREFORE conclusions to verify outcomes.
Yes, you can use a reasoning protocol for file operations and code changes. It is universally applicable to any action with potential side effects, requiring before-action planning and after-action verification to reduce missteps.
The best way to verify outcomes after an action with potential side effects is conducting an AFTER review comparing the actual RESULT against expectations to determine MATCHES and define THEREFORE next steps or STOP conditions.
You need to specify IF YES and IF NO conditions before performing an action to predefine next steps for both expected and unexpected outcomes, preventing impulsive decisions and reducing errors during tool calls or debugging.
lucaswallGenerate structured fix plans and Linear issues from bug findings.
OneWave-AICompare players side-by-side with era-adjusted statistics and advanced metric explanations.
humanerd-drewSynthesize virtual tool calls to display 'Thinking...' spinner cards in ACP clients.
alter123-zzDetect security, permission, and malicious-code risks in third-party AI skills.
