Debug
Automatically identifies and suggests fixes for common coding errors in Phrazzld projects, accelerating debugging workflows.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add phrazzld-debug npx -- -y @trustedskills/phrazzld-debug
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"phrazzld-debug": {
"command": "npx",
"args": [
"-y",
"@trustedskills/phrazzld-debug"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
What it does
This skill, phrazzld-debug, provides debugging capabilities for AI agents. It allows users to inspect and modify the agent's internal state during execution, facilitating troubleshooting and understanding of complex workflows. The tool is designed specifically for use with Claude configurations.
When to use it
- Troubleshooting unexpected behavior: If your agent isn’t performing as expected, use this skill to step through its logic and identify the root cause.
- Understanding internal state: Gain insights into how your agent processes information by observing its intermediate variables and decisions.
- Debugging complex workflows: When dealing with multi-step tasks or intricate chains of reasoning, use debugging to pinpoint bottlenecks or errors.
- Experimenting with configurations: Test different Claude configuration settings and observe their impact on the agent's behavior in real time.
Key capabilities
- Inspect internal state during execution
- Modify agent’s internal state
- Designed for Claude configurations
Example prompts
- "Debug my agent and show me the value of 'current_step'."
- "Step through the agent's process and let me examine the 'reasoning_chain' variable."
- "Pause execution and set 'task_priority' to 10."
Tips & gotchas
This skill requires a working Claude configuration. Be cautious when modifying internal state, as incorrect changes can lead to unpredictable behavior or errors.
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.