# abhishekash-mcp-trace

Source: https://glimind.com/server/pypi%2Fabhishekash-mcp-trace
Registry id: `pypi/abhishekash-mcp-trace`
Status: unknown

## Install

```json
{
  "mcpServers": {
    "abhishekash-mcp-trace": {
      "command": "uvx",
      "args": [
        "abhishekash-mcp-trace"
      ]
    }
  }
}
```

Claude Code:

```sh
claude mcp add abhishekash-mcp-trace -- uvx abhishekash-mcp-trace
```

## Tools (7)

- `approval_log`: HITL audit of approve/deny/edit decisions with rationale
- `list_runs`: Recent runs with task, model, duration, cost, decision counts
- `run_summary`: One run at a glance, accepts trace-id prefix
- `search_spans`: Find spans by tool name, file path, or other criteria
- `slowest_spans`: Top-k spans by duration to find performance issues
- `span_tree`: Nested shape of the run showing what the agent did
- `token_usage`: Aggregated token and cost usage across runs or per-run

Measured by Glimind (https://glimind.com): MCP handshake + tool list only, no tool is ever called. CC BY 4.0.