fable-ish

Risk-proportional verification adapts the depth of safety checks based on the risk level of your code tasks. By classifying tasks into 'quick', 'normal', 'deep', or 'blocked', the workflow ensures that complex deployments receive appropriate, proof-based completion tracking.

Proof-based completion ensures that only code tasks with verifiable proof of execution are considered finished. This mechanism guarantees that deployment safety checks are rigorously met before a task concludes, providing compliance reports and tracking changes.

To execute a deployment with fable-ish, use the 'fable-ish deploy' command within Claude Code. This automates the deployment process, applying deep verification and tracking changes to ensure safe, reliable code execution with verifiable proof.

Yes, fable-ish is designed for development, deployment, and maintenance tasks requiring code verification. It manages these tasks with risk-appropriate workflow steps, ensuring that even routine maintenance receives verifiable proof and compliance checks.

To implement task classification and risk management, you need Claude Code and the fable-ish plugin installed. These dependencies enforce the discipline required to classify tasks into 'quick', 'normal', 'deep', or 'blocked' categories for appropriate verification.

A code task remains 'blocked' when the risk-driven verification process determines it cannot safely proceed without meeting specific conditions. This classification prevents unsafe deployments by ensuring verifiable proof and compliance before completion.