never-again

Analyze mistakes across customization layers and implement approved systemic repairs.

Updated Mar 18, 2026
One-click install
npx skills add https://github.com/dornstein/wingman --skill never-again
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: never-again
Source: https://github.com/dornstein/wingman/tree/main/skills/never-again
Command: npx skills add https://github.com/dornstein/wingman --skill never-again

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Root-cause a mistake and implement systemic repairs across the project's customization layers so the same class of mistake doesn't recur.

Core Features & Use Cases

  • Root-cause analysis across user prompts, conversation history, and project configurations.
  • Systemic repairs across customization layers (instructions, skills, agents, memory, docs) to prevent recurrence.
  • Guided, approvable fixes that are minimal, testable, and version-controlled.

Quick Start

Describe the mistake, or say "from this session" to analyze the most recent failure in conversation history.

Frequently Asked Questions about never-again

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I perform root-cause analysis on recurring AI agent mistakes?

Root-cause analysis on AI agent mistakes surveys customization layers like instructions, memory, and docs to identify systemic factors. It proposes targeted repairs and implements version-controlled changes to prevent the same class of error from recurring.

What is the best way to harden processes after a project failure?

Harden processes after a failure by implementing systemic repairs across customization layers. Use guided, approvable fixes that are minimal and testable, enforcing version-controlled changes to add reusable guardrails for postmortem workflows.

Can I automate postmortem workflows using conversation history?

Yes, you can automate postmortem workflows using conversation history. By analyzing recent session failures, the system identifies root causes across project configurations and applies version-controlled, testable fixes to prevent future recurrence.

How do I prevent an AI agent from repeating the same mistake?

Prevent an AI agent from repeating mistakes by implementing systemic repairs across its customization layers. This involves adding reusable guardrails, updating agent instructions, and enforcing version-controlled changes after a root-cause analysis.

Does root-cause analysis support version-controlled changes for agent instructions?

Yes, root-cause analysis supports version-controlled changes for agent instructions. It enforces testable modifications across customization layers and implements them only upon approval, ensuring safe and traceable process hardening.