fable-mode

Orchestrates multi-step workflows with dependent steps, load-bearing unknowns, and verification-sensitive handoffs.

Updated Aug 30, 2026
One-click install
npx skills add https://github.com/ryanportfolio/threejs-interview-test --skill fable-mode-ryanportfolio
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: fable-mode
Source: https://github.com/ryanportfolio/threejs-interview-test/tree/main/.agents/skills/fable-mode
Command: npx skills add https://github.com/ryanportfolio/threejs-interview-test --skill fable-mode-ryanportfolio

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Complex engineering work often involves layered tasks with dependent steps, unknowns that carry the whole plan, repeated failures, or handoffs that must be verified. Without a structured workflow, these situations lead to skipped verification, self-reviewed work, and premature commits. Fable mode provides a canonical workflow for handling exactly these hard, layered problems. ## Core Features & Use Cases - Canonical Workflow Delegation: Reads and follows the canonical skill definition at .claude/skills/fable-mode/SKILL.md, resolving all script, reference, asset, and template paths from that directory. - Codex Compatibility Layer: Applies the status and exceptions defined in .agents/CODEX-SKILL-COMPATIBILITY.md and translates Claude Code tool names to Codex equivalents per the repository AGENTS.md. - Multi-Agent Verification Guardrails: Before any Task, Agent, or Workflow step, it confirms the session exposes the required multi-agent tools and forbids replacing independent review with self-review. - Use Case: When a user asks for Fable mode on a risky refactor with dependent migration steps, the skill loads the canonical workflow, checks tool availability, and enforces independent review before any commit or deploy action. ## Quick Start Ask the assistant to use Fable mode to plan and execute a layered task with dependent steps and verified handoff.

Frequently Asked Questions about fable-mode

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

FAQPage Schema
How do I use Fable mode for a complex multi-step task?▼

Invoke the skill by asking for Fable mode on your task. It reads the canonical workflow at .claude/skills/fable-mode/SKILL.md and follows it, resolving all script and reference paths from that directory.

What is the Codex adapter for Claude skills?▼

The adapter translates a canonical Claude Code skill into Codex-compatible execution. It maps Claude tool names to Codex equivalents per the repository AGENTS.md and applies exceptions listed in CODEX-SKILL-COMPATIBILITY.md.

Does Fable mode support multi-agent Task and Workflow steps?▼

Yes, but only after confirming the current Codex session exposes the required multi-agent tools. The skill explicitly forbids replacing independent review with self-review when those tools are unavailable.

Can Fable mode commit, push, or deploy automatically?▼

No. Skill invocation alone does not authorize commits, pushes, PRs, merges, deploys, migrations, or dependency installation. Those actions require an explicit user request.

When should I not use Fable mode?▼

Avoid it for simple single-step tasks without dependent steps or verification-sensitive handoffs. The workflow overhead is designed for hard layered work with load-bearing unknowns or repeated failures.