Trace
Trace reconstructs a conversation history from fragmented messages, offering context & aiding in understanding complex interactions.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add trace npx -- -y @trustedskills/trace
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"trace": {
"command": "npx",
"args": [
"-y",
"@trustedskills/trace"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
What it does
The trace skill allows you to track and understand the flow of execution within an AI agent. It provides visibility into the sequence of actions taken, including function calls and their arguments. This helps in debugging complex agent behaviors and identifying performance bottlenecks.
When to use it
- Debugging Agent Behavior: When an agent is not behaving as expected, trace can reveal the exact steps leading up to the issue.
- Performance Analysis: Identify slow or inefficient functions within your agent's workflow.
- Understanding Complex Workflows: Gain a clear understanding of how different components interact during agent execution.
- Auditing Agent Actions: Track and review the actions taken by an agent for compliance or accountability purposes.
Key capabilities
- Function call tracking
- Argument logging
- Execution sequence visualization (implied)
Example prompts
- "Enable tracing for the next 5 steps."
- "Show me a trace of the last function call."
- "Trace all calls to the 'search_web' function."
Tips & gotchas
The effectiveness of trace depends on the underlying agent framework’s ability to expose internal execution details. Ensure your agent is designed with tracing capabilities in mind for optimal results.
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.