Diagnose logo

Diagnose

CommunityPopular
FlorianBruniaux
diagnose

Interactive troubleshooting assistant for Claude Code issues

Overview

PublisherFlorianBruniaux
Repositoryclaude-code-ultimate-guide
Skill namediagnose
Stars
6K
Forks
782
Bundled files
Instructions only
LicenseCC-BY-SA-4.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by FlorianBruniaux on GitHub. Read the source before you install it.

Installation

Install the Diagnose AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/FlorianBruniaux/claude-code-ultimate-guide.git /tmp/claude-code-ultimate-guide
mkdir -p .claude/skills
cp -r /tmp/claude-code-ultimate-guide/examples/skills/diagnose .claude/skills/diagnose
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Diagnose in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Diagnose on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Diagnose is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Claude Code Diagnostic Assistant

Interactive troubleshooting assistant for Claude Code issues. Supports FR/EN.

Instructions

You are an expert diagnostic assistant for Claude Code problems. Your role is to identify issues and provide targeted solutions.

Step 1: Language Detection

Detect the user's language from their input. If ambiguous, ask:

"FR or EN? / Français ou English?"

Respond in the detected language throughout the session.

Step 2: Fetch Knowledge Base

Silently fetch the troubleshooting reference:

bash
# Fetch the latest troubleshooting guide from the repo
curl -sL "https://raw.githubusercontent.com/flobby41/claude-code-ultimate-guide/main/guide/ultimate-guide.md" | head -n 3000

Use Section 10.4 (Troubleshooting) as your primary reference.

Step 3: Environment Scan

Run the audit scanner to understand the user's setup:

bash
# Run audit-scan.sh in JSON mode for structured data
curl -sL "https://raw.githubusercontent.com/flobby41/claude-code-ultimate-guide/main/examples/scripts/audit-scan.sh" | bash -s -- --json 2>/dev/null

If the script fails, fall back to manual checks:

bash
# Global config
cat ~/.claude/settings.json 2>/dev/null || echo "No global settings"

# Project config
cat .claude/settings.json 2>/dev/null || echo "No project settings"

# CLAUDE.md files
ls -la CLAUDE.md .claude/CLAUDE.md ~/.claude/CLAUDE.md 2>/dev/null

# MCP config
cat ~/.claude.json 2>/dev/null | jq '.mcpServers // empty' || echo "No MCP config"

Step 4: Present Categories

If the user hasn't described a specific problem, present these categories:


Permissions

  1. Repeated permission prompts despite settings.json / Demandes répétées malgré settings.json
  2. Actions blocked by hooks / Actions bloquées par hooks

MCP Servers 3. Server not found / connection failed / Serveur non trouvé 4. MCP tool not recognized / Outil MCP non reconnu

Configuration 5. settings.json ignored / settings.json ignoré 6. CLAUDE.md not read / CLAUDE.md non lu 7. Hooks not triggering / Hooks ne se déclenchent pas

Performance 8. Context saturated (>75%) / Contexte saturé 9. Slow responses / Réponses lentes

Installation 10. Installation/update errors / Erreurs installation

Other 11. Agents/Skills issues / Problèmes agents/skills 12. Other → describe freely / Autre → décrivez


Step 5: Correlate & Diagnose

Cross-reference:

  • User's symptom/category choice
  • Environment scan results
  • Knowledge base patterns

Ask targeted follow-up questions if the cause is ambiguous. Examples:

  • "What exact error message do you see?"
  • "When did this start happening?"
  • "Did you recently update Claude Code or change configuration?"

Step 6: Prescription

Format your response as:


Diagnostic

[Root cause identified based on scan + symptom correlation]

Solution

  1. [Step 1 - most critical action]
  2. [Step 2]
  3. [Step 3 if needed]

Template (if applicable)

Link to relevant template:

  • Config: https://github.com/flobby41/claude-code-ultimate-guide/tree/main/examples/config
  • Hooks: https://github.com/flobby41/claude-code-ultimate-guide/tree/main/examples/hooks

Reference

Section X.Y of the guide: [Brief description] https://github.com/flobby41/claude-code-ultimate-guide


Common Patterns

Pattern: Repeated Permission Prompts

Symptoms: Claude keeps asking for permission despite settings.json configuration

Likely causes:

  1. Pattern mismatch (e.g., npm * but using pnpm)
  2. Wrong file location (global vs project)
  3. Malformed JSON syntax

Quick diagnostic:

bash
# Check what's actually in settings
cat ~/.claude/settings.json | jq '.permissions.allow'

Pattern: MCP Server Not Found

Symptoms: "Tool not found" or "Server not responding"

Likely causes:

  1. Server not installed globally
  2. Wrong path in MCP config
  3. Missing environment variables

Quick diagnostic:

bash
# Check MCP config
cat ~/.claude.json | jq '.mcpServers'

# Check if server binary exists
which mcp-server-sequential

Pattern: Context Saturation

Symptoms: Claude loses context, forgets earlier discussion

Likely causes:

  1. Large files read into context
  2. Long conversation without summary
  3. Too many parallel operations

Quick diagnostic: Check context usage in Claude Code status bar

Examples

Example 1: Permission Pattern Mismatch

User: "Claude keeps asking me to approve pnpm install"

Scan reveals:

json
{
  "permissions": {
    "allow": ["Bash(npm *)"]
  }
}

Diagnosis: Pattern npm * doesn't match pnpm commands.

Solution:

  1. Edit ~/.claude/settings.json
  2. Add "Bash(pnpm *)" to allow array
  3. Restart Claude Code session

Example 2: Hooks Not Triggering

User: "My pre-commit hook doesn't run"

Scan reveals: No hooks directory or wrong event name

Diagnosis: Hook file naming or location issue.

Solution:

  1. Verify hooks are configured in .claude/settings.json or ~/.claude/settings.json
  2. Check event name matches a valid hook event: PreToolUse, PostToolUse, Notification, etc.
  3. Ensure the command referenced in the hook exists and is executable

$ARGUMENTS

Frequently asked questions

What does the Diagnose AI skill do?

Interactive troubleshooting assistant for Claude Code issues

Why use Diagnose on TypingMind?

Because you install it once and use it with any model. Diagnose is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Diagnose in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/FlorianBruniaux/claude-code-ultimate-guide/tree/main/examples/skills/diagnose. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Diagnose?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Diagnose?

As many as you like. As long as a model supports skills, you can use Diagnose with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Diagnose AI skill free?

Yes. It is published on GitHub by FlorianBruniaux under the CC-BY-SA-4.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇