Mode Debug
Inspects internal states and execution paths of other agents to diagnose issues and improve performance.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add mode-debug npx -- -y @trustedskills/mode-debug
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"mode-debug": {
"command": "npx",
"args": [
"-y",
"@trustedskills/mode-debug"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
What it does
The mode-debug skill allows an AI agent to enter a debugging mode, providing verbose output and internal state information. This is useful for understanding the agent's reasoning process and identifying potential errors. It enables detailed logging of actions taken and intermediate results during task execution.
When to use it
- Troubleshooting complex tasks: When an agent fails unexpectedly, activate debug mode to inspect its decision-making steps.
- Understanding agent behavior: Gain insight into how the agent is interpreting instructions and interacting with tools.
- Developing new skills or prompts: Use debug mode to observe the impact of changes on the agent's performance.
- Educational purposes: Demonstrate how an AI agent approaches a problem by showing its internal thought process.
Key capabilities
- Verbose logging of actions
- Display of intermediate results
- Access to internal state variables
- Detailed error reporting
Example prompts
- "Enable debug mode and then summarize this article."
- "Run the 'calculate_total' skill in debug mode."
- "Show me the agent’s internal state while it is browsing the web."
Tips & gotchas
Debug mode significantly increases verbosity, which can impact performance. It's recommended to disable debug mode after troubleshooting or analysis is complete.
Tags
TrustedSkills Verification
Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates — what you install today is exactly what was reviewed and verified.
Security Audits
| Gen Agent Trust Hub | Pass |
| Socket | Pass |
| Snyk | Pass |
🌐 Community
Passed automated security scans.