troubleshoot — how to use troubleshoot how to use troubleshoot, troubleshoot setup guide, troubleshoot alternative, troubleshoot vs debug, what is troubleshoot, troubleshoot install, Melee AI tournament setup, Slippi netplay troubleshooting, Moltbots setup issues

v1.0.0
GitHub

About this Skill

Perfect for AI Agents like Cursor and AutoGPT needing advanced troubleshooting capabilities for Melee AI tournaments troubleshoot is a diagnostic tool for identifying and resolving common setup issues in Melee AI tournaments, ensuring seamless gameplay and competition.

Features

Runs system checks using uname and python3.12 commands
Verifies Python version and virtual environment (venv) existence
Checks venv Python version using .venv/bin/python command
Provides step-by-step guide for diagnosing and fixing setup problems
Supports Slippi netplay and Moltbots pluggable fighters

# Core Topics

ScavieFae ScavieFae
[8]
[0]
Updated: 2/24/2026

Quality Score

Top 5%
33
Excellent
Based on code quality & docs
Installation
SYS Universal Install (Auto-Detect)
Cursor IDE Windsurf IDE VS Code IDE
> npx killer-skills add ScavieFae/nojohns/troubleshoot

Agent Capability Analysis

The troubleshoot MCP Server by ScavieFae is an open-source Community integration for Claude and other AI agents, enabling seamless task automation and capability expansion. Optimized for how to use troubleshoot, troubleshoot setup guide, troubleshoot alternative.

Ideal Agent Persona

Perfect for AI Agents like Cursor and AutoGPT needing advanced troubleshooting capabilities for Melee AI tournaments

Core Value

Empowers agents to diagnose and fix common setup problems using system checks, Python version validation, and venv existence verification, all through Unix-based commands like uname and python3

Capabilities Granted for troubleshoot MCP Server

Debugging Slippi netplay issues
Resolving Moltbots setup problems
Validating Python version for compatibility

! Prerequisites & Limits

  • Requires Unix-based system for command execution
  • Python 3.12.x required, not compatible with Python 3.13
  • Venv setup necessary for proper functionality
Project
SKILL.md
6.2 KB
.cursorrules
1.2 KB
package.json
240 B
Ready
UTF-8

# Tags

[No tags]
SKILL.md
Readonly

Setup Troubleshooter

Diagnose and fix common No Johns setup problems. Run each check in order and report findings at the end.

Steps

1. System Check

bash
1# OS and architecture 2uname -a 3 4# Python version (must be 3.12.x, NOT 3.13) 5python3.12 --version 2>/dev/null || python3 --version 6 7# Check if venv exists 8ls -la .venv/bin/python 2>/dev/null && echo "Venv OK" || echo "PROBLEM: No .venv found" 9 10# Check venv Python version 11.venv/bin/python --version 2>/dev/null

Expected: Python 3.12.x. If 3.13, pyenet won't build. If no venv, user needs python3.12 -m venv .venv.

2. Package Check

bash
1# libmelee (critical — the Dolphin interface) 2.venv/bin/python -c "import melee; print('libmelee OK')" 2>&1 3 4# pyenet (libmelee dependency — C extension, common build failure) 5.venv/bin/python -c "import enet; print('pyenet OK')" 2>&1 6 7# nojohns CLI 8.venv/bin/python -m nojohns.cli list-fighters 2>&1 9 10# TensorFlow (only needed for Phillip) 11.venv/bin/python -c "import tensorflow as tf; print(f'TF {tf.__version__} OK')" 2>&1 12 13# slippi-ai (Phillip runtime) 14.venv/bin/python -c "import slippi_ai; print('slippi-ai OK')" 2>&1

Common failures:

  • import enet fails → pyenet build issue. See docs/TROUBLESHOOTING.md "pyenet Build Failure"
    • macOS fix: LDFLAGS="-L/opt/homebrew/lib -lenet" CFLAGS="-I/opt/homebrew/include" .venv/bin/pip install --no-cache-dir --no-binary :all: pyenet
    • Linux fix: sudo apt install -y libenet-dev && .venv/bin/pip install --force-reinstall pyenet
  • import melee fails → libmelee not installed. Run .venv/bin/pip install -e .
  • import tensorflow fails with mutex lock → TF 2.20 on macOS ARM. Fix: .venv/bin/pip install "tensorflow==2.18.1" "tf-keras==2.18.0"

3. Config Check

bash
1# Config file exists 2cat ~/.nojohns/config.toml 2>/dev/null || echo "PROBLEM: No config.toml — run 'nojohns setup melee'"

Verify these fields in ~/.nojohns/config.toml:

  • dolphin_path: Must exist and contain "netplay" in the path string (libmelee validates this)
    • macOS: ~/Library/Application Support/Slippi Launcher/netplay
    • Linux: ~/.config/Slippi Launcher/netplay
  • iso_path: Must exist and be >500MB (a real Melee ISO is ~1.3GB)
  • connect_code: Format XXXX#NNN (4 letters, hash, 1-3 digits). Avoid codes with '9' on macOS Sequoia.
bash
1# Verify Dolphin path exists 2DOLPHIN_PATH=$(python3 -c " 3import tomllib 4with open('$HOME/.nojohns/config.toml', 'rb') as f: 5 c = tomllib.load(f) 6print(c.get('games', {}).get('melee', {}).get('dolphin_path', 'NOT SET')) 7" 2>/dev/null) 8echo "Dolphin path: $DOLPHIN_PATH" 9ls "$DOLPHIN_PATH" 2>/dev/null && echo "Path OK" || echo "PROBLEM: Path does not exist" 10 11# Verify ISO exists and check size 12ISO_PATH=$(python3 -c " 13import tomllib 14with open('$HOME/.nojohns/config.toml', 'rb') as f: 15 c = tomllib.load(f) 16print(c.get('games', {}).get('melee', {}).get('iso_path', 'NOT SET')) 17" 2>/dev/null) 18echo "ISO path: $ISO_PATH" 19ls -la "$ISO_PATH" 2>/dev/null || echo "PROBLEM: ISO file not found"

4. Dolphin Check

bash
1# macOS: Check Gatekeeper isn't blocking Dolphin 2# (only relevant on macOS) 3if [[ "$(uname)" == "Darwin" ]]; then 4 DOLPHIN_APP="$HOME/Library/Application Support/Slippi Launcher/netplay/Slippi Dolphin.app" 5 if [ -d "$DOLPHIN_APP" ]; then 6 echo "Dolphin app found" 7 xattr -l "$DOLPHIN_APP" 2>/dev/null | grep -q quarantine && \ 8 echo "PROBLEM: Gatekeeper quarantine active. Fix: xattr -cr \"$DOLPHIN_APP\"" || \ 9 echo "Gatekeeper OK" 10 else 11 echo "PROBLEM: Dolphin.app not found at expected location" 12 echo "Open Slippi Launcher and let it download Dolphin" 13 fi 14fi 15 16# Linux: Check Dolphin binary exists 17if [[ "$(uname)" == "Linux" ]]; then 18 NETPLAY_DIR="$HOME/.config/Slippi Launcher/netplay" 19 if [ -d "$NETPLAY_DIR" ]; then 20 echo "Slippi netplay dir found" 21 find "$NETPLAY_DIR" -name "dolphin-emu" -o -name "*.AppImage" 2>/dev/null 22 else 23 echo "PROBLEM: Slippi netplay directory not found" 24 fi 25fi

5. Phillip Check

bash
1# Model weights exist 2MODEL="fighters/phillip/models/all_d21_imitation_v3.pkl" 3if [ -f "$MODEL" ]; then 4 SIZE=$(wc -c < "$MODEL") 5 echo "Model weights: $SIZE bytes" 6 if [ "$SIZE" -lt 1000000 ]; then 7 echo "PROBLEM: Model file too small — may be corrupted" 8 else 9 echo "Model OK" 10 fi 11else 12 echo "PROBLEM: Model weights not found. Run: nojohns setup melee phillip" 13fi 14 15# slippi-ai installed 16if [ -d "fighters/phillip/slippi-ai" ]; then 17 echo "slippi-ai repo OK" 18else 19 echo "PROBLEM: slippi-ai not cloned. Run: nojohns setup melee phillip" 20fi

6. Network Check

bash
1# Arena reachable 2ARENA_URL=$(python3 -c " 3import tomllib 4with open('$HOME/.nojohns/config.toml', 'rb') as f: 5 c = tomllib.load(f) 6print(c.get('arena', {}).get('url', 'https://nojohns-arena-production.up.railway.app')) 7" 2>/dev/null || echo "https://nojohns-arena-production.up.railway.app") 8 9echo "Arena URL: $ARENA_URL" 10curl -s --max-time 5 "$ARENA_URL/health" 2>/dev/null || echo "PROBLEM: Arena unreachable"

7. Known Issues Reference

If the checks above pass but things still don't work, check these known issues in docs/TROUBLESHOOTING.md:

SymptomIssueFix
Connect codes with '9' fail on SequoiaPosition 3 bugUse a code without '9'
Netplay freezes at stock lossRollback desyncSet online_delay=6
Phillip doesn't move in netplayMissing on_game_startUpdate to latest code
mutex lock failed on TF importTF 2.20 bug on ARMInstall TF 2.18.1
Unknown path from libmeleePath missing "netplay"Use Slippi Launcher's netplay dir
Menu stuck on 2nd+ matchlibmelee state leakUse subprocess per match
CSS stuck at character selectIntermittentKill and restart matchmake

8. Report

After running all checks, print a summary:

=== No Johns Setup Diagnostic ===
System:     [OS, arch, Python version]
Packages:   [OK / PROBLEMS: list]
Config:     [OK / PROBLEMS: list]
Dolphin:    [OK / PROBLEMS: list]
Phillip:    [OK / PROBLEMS: list]
Network:    [OK / PROBLEMS: list]

[For each PROBLEM, include the specific fix command]

If everything passes: "Setup looks good. Try nojohns fight random do-nothing for a smoke test."

Related Skills

Looking for an alternative to troubleshoot or building a Community AI Agent? Explore these related open-source MCP Servers.

View All

widget-generator

Logo of f
f

widget-generator is an open-source AI agent skill for creating widget plugins that are injected into prompt feeds on prompts.chat. It supports two rendering modes: standard prompt widgets using default PromptCard styling and custom render widgets built as full React components.

149.6k
0
Design

testing

Logo of lobehub
lobehub

Testing is a process for verifying AI agent functionality using commands like bunx vitest run and optimizing workflows with targeted test runs.

73.3k
0
Communication

chat-sdk

Logo of lobehub
lobehub

chat-sdk is a unified TypeScript SDK for building chat bots across multiple platforms, providing a single interface for deploying bot logic.

73.0k
0
Communication

zustand

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication