capability-upgrade

Construct and upgrade a Capability Pack with staged gate-based quality checks.

3|Updated Sep 27, 2025
One-click install
npx skills add https://github.com/Sheldon-92/TAD --skill capability-upgrade
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: capability-upgrade
Source: https://github.com/Sheldon-92/TAD/tree/main/.agents/skills/capability-upgrade
Command: npx skills add https://github.com/Sheldon-92/TAD --skill capability-upgrade

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Capability Pack 构建/升级流程从评估到研究再到构建,提供完整方法论,基于 2026-05-07 web-ui-design 升级实战经验,且 YAML Domain Packs 已于 2026-06-11 停止维护,归档源可在 .tad/archive/domains/ 查阅。

Core Features & Use Cases

  • Produces a standalone capability pack directory (CAPABILITY.md) with an entry protocol and N capabilities, plus install.sh, checklists, tools, references, and examples.
  • Standardizes end-to-end lifecycle: evaluation, research, design, build, and validation for domain capabilities.
  • Enables reuse across projects by capturing anti-slop rules, gating/validation, and modular components for on-demand execution.

Quick Start

Run the end-to-end capability-pack workflow to evaluate a domain, conduct targeted research, design the pack, and generate CAPABILITY.md along with the supporting files.

Frequently Asked Questions about capability-upgrade

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

FAQPage Schema
How do I build a capability pack for AI agents from scratch?

To build a capability pack for AI agents, run the end-to-end workflow covering evaluation, research, design, and validation to generate a standalone directory with CAPABILITY.md, install.sh, checklists, tools, references, and examples.

What is a standardized workflow for upgrading domain-specific AI agent capabilities?

A standardized capability upgrade workflow applies gate-based quality checks across multiple lifecycle stages—evaluation, research, design, build, and validation—to ensure domain-specific AI agent capabilities meet anti-slop quality bars before deployment.

Can I reuse a capability pack across multiple projects with different AI agents?

Yes, capability packs are designed for reuse across projects by capturing modular components, anti-slop rules, and gating validations for on-demand execution within different domain-specific AI agent teams.

What files and artifacts are required for a complete AI agent capability pack?

A complete AI agent capability pack requires CAPABILITY.md with an entry protocol, install.sh, checklists, tools, references, and examples, providing explicit CLI commands and staged validation gates.

Does the capability pack workflow enforce quality checks during the build process?

Yes, the capability pack workflow enforces quality through gate-based validation checks staged across evaluation, research, design, build, and validation, ensuring anti-slop rules are met before completing the pack.

When should I upgrade an existing domain capability pack instead of building a new one?

Upgrade an existing domain capability pack when applying new design patterns or validation steps from standardized workflows, especially since older formats like YAML Domain Packs are archived and no longer maintained.