More screenshots
Works with the AI tools you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+22 moreClaude CodeCursorCodex CLIMuseDots
OpenClaw
Manus
Grok
Grok Bot+18 agentsAgentTrace MCP: Trace runs. Guard schemas. Catch regressions
Audit agent trajectories, validate MCP schemas, and detect tool-call regressions with a professional dashboard.
$9.99
AgentTrace MCP: Trace runs. Guard schemas. Catch regressions
Example session with this skill installed
Use AgentTrace MCP in Guard mode only (do not run a full trajectory audit).
Validate these MCP tool schemas and the following proposed tool call. Use strict mode.
- Read your context and instructions
- Compiled the agenttrace mcp
Connects securely to your tools. The creator never sees your data.
About this skill
Debug Agents. Guard Tools. Catch Regressions.
Turn agent traces and MCP logs into clear, evidence-based reliability reports. Find failures, validate schemas, detect silent corruption and compare runs.
🧠 About AgentTrace
AgentTrace MCP is an AI agent reliability skill for analyzing tool-using agents, MCP workflows and multi-step runs. It transforms raw execution data into structured findings and actionable fixes.
⚙️ Four powerful modes
🔍 Audit — reconstruct and analyze agent runs.
🛡️ Guard — validate MCP schemas and tool calls.
📈 Regression — compare runs against baselines.
📦 Batch — detect recurring failures across runs.
🔥 Core capabilities
🔍 Trace Reconstruction
Turn raw logs and tool calls into a readable execution timeline.
🛡️ MCP Schema Validation
Find invalid arguments, missing fields, type errors and schema drift.
🚨 Failure Detection
Identify the first critical failure instead of only reporting the final error.
🕵️ Silent Corruption Detection
Catch subtle data, type, argument and tool-call problems.
📊 Regression Analysis
Compare current behavior with known-good baselines.
🎯 Perfect for
🤖 Agent debugging • 🛡️ MCP validation • 📈 Regression testing • 🔬 Reliability audits • 📦 Batch analysis • ⚙️ CI workflows
📊 Clear engineering reports
Get structured findings, metrics, risks, evidence and recommended actions in HTML, Markdown or JSON formats.
🔍 🛡️ 📈Know why your agent failed.
Trace it. Validate it. Fix it. Prevent it from happening again.
How to install
Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.
- 1
Download the ZIP
Free skills download straight away. Paid skills unlock right after purchase.
- 2
Unzip into your skills folder
Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.
- 3
Ask your agent to use it
Restart the agent if it was already running. It picks the skill up automatically - no config needed.
Skills folder by agent
Click the path to copy it. Create the folder if it does not exist yet.
Reviews
2 people have installed this skill.
Trust & safety
Security scanned
Verified clean 1 month ago
- Passed all security checks, Safe to install