harden-task-file

Harden /define task guidance files with explicit acceptance criteria and convergence checks.

70|9|Updated Jan 27, 2026
One-click install
npx skills add https://github.com/doodledood/manifest-dev --skill harden-task-file
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: harden-task-file
Source: https://github.com/doodledood/manifest-dev/tree/main/.claude/skills/harden-task-file
Command: npx skills add https://github.com/doodledood/manifest-dev --skill harden-task-file

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Harden /define task guidance files to ensure one-shot quality by enforcing comprehensive coverage, explicit acceptance criteria, and a structured iteration process.

Core Features & Use Cases

  • Orthogonality analysis: map every item to independent dimensions to prevent redundancy.
  • Iteration loop: log rounds, propose user-approved additions, and run prompt reviews to converge.
  • Convergence guarantees: enforce per-round checks and a final validation to ensure no missing dimensions.
  • Reusability across domains: apply to any /define task file to tighten quality gates, risks, scenarios, and trade-offs.

Quick Start

Provide the target /define task guidance file and let this skill iteratively harden it until all acceptance criteria are explicit and coverage is complete.

Frequently Asked Questions about harden-task-file

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

FAQPage Schema
How do I harden task files to ensure one-shot prompt quality?

To harden task files for one-shot prompt quality, apply an iterative process enforcing orthogonality analysis, explicit acceptance criteria, and convergence checks. This guarantees complete coverage and prevents missing dimensions in task guidance files.

What is orthogonality analysis in task guidance file iteration?

Orthogonality analysis maps every task file item to independent dimensions to prevent redundancy. It ensures each risk, scenario, and trade-off section remains distinct, tightening quality gates and eliminating overlapping definitions.

How do I enforce convergence checks during prompt engineering iterations?

Enforce convergence checks by logging per-round iterations, proposing user-approved additions, and running prompt reviews. A final validation ensures no missing dimensions remain, guaranteeing the task file meets complete coverage acceptance criteria.

Can I apply task file hardening to any domain-specific manifest?

Yes, you can apply task file hardening to any domain-specific manifest. It reusably tightens quality gates, risks, scenarios, and trade-offs across diverse domains by iteratively refining explicit acceptance criteria until coverage is complete.

What's the best way to define explicit acceptance criteria in task files?

The best way to define explicit acceptance criteria is through a structured iteration loop proposing user-approved additions and running prompt reviews. This enforces per-round logging and validates that all coverage dimensions are fully satisfied.