About Agent Trace Check
Agent Trace Check is a read-only MCP server for examining AI agent tool-call logs. It helps developers identify failed, repeated, or slow calls; the README lists it among Openkrill's public developer and security checks.
Key features
- 01Inspect agent tool-call logs
- 02Identify failed calls
- 03Find repeated calls
- 04Spot slow calls
- 05Public access without an API key
Use cases
- 01Debugging failures in an agent run
- 02Finding loops in tool-call sequences
- 03Reviewing slow tool calls
MCP support
It exposes Agent Trace Check through the MCP endpoint https://agenttrace.openkrill.app/mcp. The server is public, read-only, and requires no key.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://agenttrace.openkrill.app/mcpclaude mcp add --transport http app-openkrill-agent-trace https://agenttrace.openkrill.app/mcp{"mcpServers":{"app-openkrill-agent-trace":{"type":"http","url":"https://agenttrace.openkrill.app/mcp"}}}Open in Cursor{"servers":{"app-openkrill-agent-trace":{"type":"http","url":"https://agenttrace.openkrill.app/mcp"}}}Open in VS Code{"mcpServers":{"app-openkrill-agent-trace":{"type":"http","url":"https://agenttrace.openkrill.app/mcp"}}}{"mcpServers":{"app-openkrill-agent-trace":{"serverUrl":"https://agenttrace.openkrill.app/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
3| Parameter | Type | Required | Description |
|---|---|---|---|
| trace | string | Yes | The tool-call log as text: JSON lines or one JSON value. Reads OpenAI chat messages with tool_calls, Responses API function_call items, Anthropic tool_use and tool_result blocks, Claude Code session JSONL, MCP tools/call requests and answers, LangSmith tool runs, OpenTelemetry tool spans, and flat rows with tool, args, status or error, and duration_ms or timestamps. |
| slow_ms | integer | No | A call at or over this many milliseconds is slow. Default 10000. |
| retry_min | integer | No | Back-to-back calls to one tool that count as a retry loop or a repeated call. Default 3. |
Details
From the project READMEDeveloper and Security Checks
Package health, CVE triage, software end-of-life and licence facts, DESIGN.md lint and diff, and AI coding agent news.
claude plugin install dev-security@openkrill
| Skill and MCP server | Tool | Endpoint |
|---|---|---|
packages | Package Health Check | https://packages.openkrill.app/mcp |
cve | CVE Risk Check | https://cve.openkrill.app/mcp |
dev-facts | Dev Facts Check | https://devfacts.openkrill.app/mcp |
design-md | DESIGN.md Linter & Extractor | https://designmd.openkrill.app/mcp |
industry-pulse | Industry Pulse | https://pulse.openkrill.app/mcp |
agent-trace | Agent Trace Check | https://agenttrace.openkrill.app/mcp |
injection-scan | Prompt Injection Scanner | https://injectionscan.openkrill.app/mcp |
Every server is public, read-only and needs no key. Generated from the Agent Tools plugin listings; edits made here are overwritten.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.