A Claude skill that enforces staged execution discipline on large tasks, enabling explicit multi-stage planning, sub-agent delegation, and self-verification. It is designed for users tackling complex, multi-part tasks requiring systematic execution and robust verification, aiming to impose structure and enhance reliability.

How It Works

The skill shapes model procedure by enforcing decomposition before action, delegating independent sub-work via sub-agent tooling, verifying each stage with failable checks, and critiquing output before delivery. This approach imposes structure and enhances reliability for complex tasks, especially on weaker models, by reinforcing systematic execution rather than increasing raw model capability.

Quick Start & Requirements

• Installation: Place the fable-mode directory wherever your Claude environment loads skills from (e.g., a skills directory read by Claude Code).
• Prerequisites: None explicitly mentioned beyond a Claude environment capable of loading skills.
• Links: SKILL.md, EXAMPLE.md

Highlighted Details

• Enforces explicit multi-stage planning.
• Supports parallel delegation where runtime allows.
• Includes a verification check at each stage that can actually fail.
• Features a skeptical self-review before delivery.

Maintenance & Community

No information on contributors, sponsorships, community links, or roadmap is provided in the README.

Licensing & Compatibility

The README does not specify a license or compatibility notes for commercial use.

Limitations & Caveats

The skill does not enhance the underlying model's inherent reasoning capabilities; it imposes structure but cannot raise the reasoning ceiling. It is not intended for trivial tasks that fit within a single pass, as staging them would be inefficient. The skill is described as a "checklist, not a capability transplant."