agentdev-workflow-inspect-skills

Diagnoses Command and Skill definitions for structural validity and outputs findings to inbox files.

Updated Mar 7, 2026
One-click install
npx skills add https://github.com/yogata/agent-dev-flow --skill agentdev-workflow-inspect-skills-yogata
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agentdev-workflow-inspect-skills
Source: https://github.com/yogata/agent-dev-flow/tree/main/src/opencode/skills/agentdev-workflow-inspect-skills
Command: npx skills add https://github.com/yogata/agent-dev-flow --skill agentdev-workflow-inspect-skills-yogata

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? It detects structural and semantic problems in Command and Skill definition files—such as broken Command→Skill references, invalid frontmatter, poor granularity, responsibility boundary violations, and Markdown syntax corruption—without modifying the diagnosed files, and records findings for later triage. ## Core Features & Use Cases - Semantic diagnosis of Skill structure: Evaluates reference validity, frontmatter, body structure, progressive disclosure, responsibility boundaries, canonical naming, and internal structure dependencies. - Distribution syntax health checks: Detects frontmatter duplication, duplicate headings, broken Markdown, references to nonexistent commands, encoding inconsistencies (UTF-8 BOM, mixed CRLF/LF), and responsibility description contradictions. - Finding classification and routing: Classifies each finding (false positive, pre-existing, fix target), presents recommended routes without auto-fixing, writes findings to .agentdev/inspect/inbox/, and persists them via git commit/push. - Use Case: Run the /agentdev/inspect-skills command after editing Skill definitions to receive a classified diagnostic report of structural issues, persisted as finding files for the inspect-promote workflow to triage. ## Quick Start Run the /agentdev/inspect-skills command to diagnose all Command and Skill definitions and generate a classified findings report.

Frequently Asked Questions about agentdev-workflow-inspect-skills

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

FAQPage Schema
How do I diagnose Skill definition files for structural problems?▼

Run the /agentdev/inspect-skills command, which dispatches to this workflow skill. It reads Command and Skill definitions, evaluates reference validity, frontmatter, granularity, and responsibility boundaries, then outputs a classified findings report.

What does the inspect-skills diagnostic check in Skill frontmatter?▼

It validates frontmatter structure, canonical naming, progressive disclosure, responsibility boundaries, and internal structure dependencies. It also checks distribution files for duplicate frontmatter, duplicate headings, broken Markdown, and encoding inconsistencies like UTF-8 BOM or mixed line endings.

Does inspect-skills automatically fix the problems it finds?▼

No, it is a read-only diagnostic type that never modifies diagnosed files. It only classifies findings, presents recommended fix routes, and writes finding files to .agentdev/inspect/inbox/ for the inspect-promote workflow to triage.

Can I run the inspect-skills workflow skill directly without the command?▼

No, it must be invoked through the corresponding /agentdev/inspect-skills command. The command owns the public interface, guardrails, and dispatch, while this skill provides only the workflow implementation.

Why does inspect-skills report a reference to a nonexistent command?▼

It cross-checks README listings and command bodies for references pointing to commands that do not exist, flagging them as findings per the syntax integrity inspection rules. References to existing commands are not reported.