fable-mode

Enforce staged planning, parallel delegation, and verification for complex multi-step tasks.

12|3|Updated Apr 6, 2026
One-click install
npx skills add https://github.com/tsukhani/jclaw --skill fable-mode-tsukhani
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: fable-mode
Source: https://github.com/tsukhani/jclaw/tree/main/skills/fable-mode
Command: npx skills add https://github.com/tsukhani/jclaw --skill fable-mode-tsukhani

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the problem of cognitive drift and oversight in complex, multi-step projects by enforcing a structured, verifiable, and self-critical execution loop.

Core Features & Use Cases

  • Staged Decomposition: Forces the creation of a formal stage map before any action is taken, ensuring clear objectives.
  • Verification Discipline: Mandates failable checks for every stage, preventing the propagation of errors in long-running tasks.
  • Self-Critique: Requires a skeptical review of all outputs before final delivery to identify weaknesses.
  • Use Case: Use this when performing deep-work tasks like refactoring a large codebase, conducting multi-source research, or managing long-running data analysis projects that span multiple sessions.

Quick Start

Activate fable-mode to systematically plan and execute the deep-work task of refactoring the authentication module.

Frequently Asked Questions about fable-mode

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

FAQPage Schema
How do I prevent errors when executing complex, multi-step tasks?

To prevent errors in complex tasks, you need a disciplined execution framework that mandates failable checks for every stage. This enforces rigorous verification at each step, preventing error propagation in long-running workflows.

What is the best way to manage a large codebase refactoring workflow?

The best way to manage a large codebase refactoring workflow is through staged decomposition, which forces the creation of a formal stage map before taking action. This ensures clear objectives and systematic execution across multiple sessions.

How does parallel delegation work for multi-source research tasks?

Parallel delegation for multi-source research works by coordinating concurrent processes through subagent tools. This manages state and validates outputs simultaneously, ensuring high coherence and systematic execution across diverse data sources.

Do I need filesystem and exec tools to coordinate concurrent processes?

Yes, you need filesystem and exec tools to coordinate concurrent processes for complex workflows. These tools are required to manage state, validate outputs, and execute parallel delegation across multiple concurrent subagents.

Why does cognitive drift happen in long-running data analysis projects?

Cognitive drift happens in long-running data analysis projects due to a lack of self-critique and structured verification. Without a skeptical review of outputs before final delivery, weaknesses and oversight accumulate across extended sessions.