paishu-ai/pai-zong-nb-v5.8/28-diagnose-pro

Claude Code and agent skill by paishu-ai/pai-zong-nb-v5.8

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Disciplined debugging and diagnosis workflow for hard bugs, regressions, flaky behavior, performance issues, and unknown failures. Adapted from Matt Pocock-style diagnosis loops for Codex v4.1. Use when root cause is unclear or the bug is not fixed by a direct local change.

Install

npx vskill@latest install paishu-ai/pai-zong-nb-v5.8/28-diagnose-pro

SKILL.md

The same SKILL.md was published earlier as paishu-ai/pai-zong-nb-v5.8/28-diagnose-pro.

Diagnose Pro

Workflow

  1. Reproduce or capture the failure as concretely as possible.
  2. Reduce the problem to the smallest failing path.
  3. Form one hypothesis at a time.
  4. Add observation: logs, assertions, tests, traces, screenshots, query plans, or minimal probes.
  5. Fix the root cause, not the symptom.
  6. Add a regression test or verification step.
  7. Remove temporary probes before finishing unless they are useful permanent diagnostics.

Diagnosis Rules

  • Prefer evidence over intuition.
  • Change only one variable per diagnostic step when possible.
  • Treat "works on my machine" as an environment-difference bug, not a conclusion.
  • For flaky bugs, collect timing, concurrency, cache, external API, and environment signals.
  • For performance bugs, establish a baseline metric before optimizing.
  • For security bugs, preserve exploit conditions and prove the boundary is closed.

Output

Use this shape:

markdown
根因:...
证据:...
修复:...
验证:...

If root cause is still unknown, state the strongest remaining hypothesis and the next concrete observation needed.

Boundaries

  • Do not claim files, generated media, deployments, scans, account changes, or external writes unless the active tool or command output proves them.
  • Do not override system, developer, user, project, or selected-skill rules.
  • If another skill is more specific for the user's intent, hand off or use it as the main skill instead of duplicating its workflow.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
1
Repository forks
1
7-day trend score
8

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/paishu-ai/pai-zong-nb-v5.8/28-diagnose-pro/badge)](https://verified-skill.com/skills/paishu-ai/pai-zong-nb-v5.8/28-diagnose-pro)

Something wrong with this skill or its page? Tell us.