Skip to content
Get packs

AI Workflow Diagnostics

Audits an AI agent workflow across 5 quality dimensions and returns a scored report with fixes.

Source Download skill
Repository
github/awesome-copilot
Stars
40,703
Licence
MIT
Files
1 file

What it does

When you ask the agent to diagnose a workflow, it reviews your AI setup (prompts, tools, context handling, architecture, and safety) and scores each area from 1 to 5. It then lists the most serious problems and gives a prioritized list of concrete fixes. It works from the description, prompt text or tool list you paste in, and does not run any code or connect to outside services.

Great for

  • Developers who have built an AI agent or workflow and want a review before production
  • Teams doing a health check after changing prompts, tools or agent setup
  • People who want a prioritized remediation plan rather than a general critique

Skip it if

  • Auditing a live system's real behaviour: the skill only reviews the text and configuration you provide, so it cannot confirm how the workflow actually performs (reason: scores are judgement-based from the description)
  • Replacing formal security testing: the safety dimension is a checklist-style review, not a penetration test or compliance audit

Try asking

  • “Diagnose this agent prompt and tool list before I ship it to production”
  • “Run a workflow diagnostic on my multi-agent setup and give me a prioritized fix list”
  • “Health-check my agent after I added five new tools and a longer system prompt”
Step-by-step guide: how to use AI Workflow DiagnosticsInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the MIT licence and an attribution note.

    diagnose.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip diagnose.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip diagnose.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Paste the full system prompt, the tool definitions and a short description of the workflow so the scores are grounded in real detail
  • Ask for a re-scan after applying fixes to see which scores improved
  • Treat the scores as a structured starting point for review, not an objective measurement

What's inside

  • SKILL.md

Redistributed unmodified from github/awesome-copilot under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.