Skip to main content
Glama
bmdhodl

Agentguard47

by bmdhodl
OverviewSchemaRelated ServersScoreDiscussions
Python
Local

get_trace

Retrieve the complete event tree for a given trace, including all spans, tool calls, LLM calls, guard triggers, and errors. Ideal for debugging agent runs and inspecting guard responses.

Instructions

Get the full event tree for a specific trace by its trace ID. Shows all spans, tool calls, LLM calls, guard triggers, and errors.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
trace_idYesThe trace ID to look up

Implementation Reference

  • mcp-server/src/tools.ts:1-177 (registration)
    The tools array definition - 'get_trace' is registered as one of the tools in the array (line 43) and exported for use by the MCP server index.ts which iterates and registers via server.tool().
    import { AgentGuardClient } from "./client.js";
import { extractDecisionEvents } from "./decisions.js";

export interface ToolDefinition {
  name: string;
  description: string;
  inputSchema: {
    type: "object";
    properties: Record<string, unknown>;
    required?: string[];
  };
  handler: (client: AgentGuardClient, args: Record<string, unknown>) => Promise<string>;
}

export const tools: ToolDefinition[] = [
  {
    name: "query_traces",
    description:
      "Search recent traces from your AgentGuard-instrumented agents. " +
      "Filter by service name, time range, or paginate through results.",
    inputSchema: {
      type: "object",
      properties: {
        limit: { type: "number", description: "Max traces to return (default 20, max 500)" },
        offset: { type: "number", description: "Offset for pagination" },
        service: { type: "string", description: "Filter by service name" },
        since: { type: "string", description: "ISO timestamp — only traces after this time" },
        until: { type: "string", description: "ISO timestamp — only traces before this time" },
      },
    },
    handler: async (client, args) => {
      const result = await client.getTraces({
        limit: args.limit ? String(args.limit) : "20",
        offset: args.offset ? String(args.offset) : undefined,
        service: args.service as string | undefined,
        since: args.since as string | undefined,
        until: args.until as string | undefined,
      });
      return JSON.stringify(result, null, 2);
    },
  },
  {
    name: "get_trace",
    description:
      "Get the full event tree for a specific trace by its trace ID. " +
      "Shows all spans, tool calls, LLM calls, guard triggers, and errors.",
    inputSchema: {
      type: "object",
      properties: {
        trace_id: { type: "string", description: "The trace ID to look up" },
      },
      required: ["trace_id"],
    },
    handler: async (client, args) => {
      const result = await client.getTrace(args.trace_id as string);
      return JSON.stringify(result, null, 2);
    },
  },
  {
    name: "get_trace_decisions",
    description:
      "Extract normalized decision.* events from one trace. " +
      "Use this when a workflow includes proposal, override, approval, or binding steps.",
    inputSchema: {
      type: "object",
      properties: {
        trace_id: { type: "string", description: "The trace ID to inspect for decision events" },
      },
      required: ["trace_id"],
    },
    handler: async (client, args) => {
      const traceId = args.trace_id as string;
      const result = await client.getTrace(traceId);
      const events = Array.isArray(result.events) ? result.events : [];
      const decisions = extractDecisionEvents(events as Array<Record<string, unknown>>, {
        traceId,
      });
      return JSON.stringify({ trace_id: traceId, decisions }, null, 2);
    },
  },
  {
    name: "get_alerts",
    description:
      "Get recent guard alerts (loop detection, budget exceeded) and errors. " +
      "Useful for checking if your agents are hitting safety limits.",
    inputSchema: {
      type: "object",
      properties: {
        limit: { type: "number", description: "Max alerts to return (default 50)" },
        since: { type: "string", description: "ISO timestamp — only alerts after this time" },
      },
    },
    handler: async (client, args) => {
      const result = await client.getAlerts({
        limit: args.limit ? String(args.limit) : undefined,
        since: args.since as string | undefined,
      });
      return JSON.stringify(result, null, 2);
    },
  },
  {
    name: "get_usage",
    description:
      "Check your current event quota usage and plan limits. " +
      "Shows event count vs limit, retention period, and plan details.",
    inputSchema: {
      type: "object",
      properties: {},
    },
    handler: async (client) => {
      const result = await client.getUsage();
      const pct = result.event_limit > 0
        ? ((result.event_count / result.event_limit) * 100).toFixed(1)
        : "0";
      return JSON.stringify({ ...result, usage_percent: `${pct}%` }, null, 2);
    },
  },
  {
    name: "get_costs",
    description:
      "Get cost breakdown for the current month: total spend, cost by model, " +
      "and estimated savings from guard interventions.",
    inputSchema: {
      type: "object",
      properties: {},
    },
    handler: async (client) => {
      const result = await client.getCosts();
      return JSON.stringify(result, null, 2);
    },
  },
  {
    name: "check_budget",
    description:
      "Quick pass/fail budget health check. Combines usage quota and cost data " +
      "to give a summary of whether you're within safe operating limits.",
    inputSchema: {
      type: "object",
      properties: {},
    },
    handler: async (client) => {
      const [usage, costs] = await Promise.all([
        client.getUsage(),
        client.getCosts(),
      ]);

      const usagePct = usage.event_limit > 0
        ? (usage.event_count / usage.event_limit) * 100
        : 0;

      const status = usagePct >= 90
        ? "critical"
        : usagePct >= 75
          ? "warning"
          : "healthy";

      return JSON.stringify(
        {
          status,
          plan: usage.plan,
          events: {
            used: usage.event_count,
            limit: usage.event_limit,
            percent: `${usagePct.toFixed(1)}%`,
          },
          costs: {
            monthly_total: costs.monthly.total_cost,
            trace_count: costs.monthly.trace_count,
          },
          savings: costs.savings,
        },
        null,
        2,
      );
    },
  },
];
  • mcp-server/src/tools.ts:42-57 (handler)
    The 'get_trace' tool handler function. It calls client.getTrace(args.trace_id) and returns the result as a JSON string.
    {
  name: "get_trace",
  description:
    "Get the full event tree for a specific trace by its trace ID. " +
    "Shows all spans, tool calls, LLM calls, guard triggers, and errors.",
  inputSchema: {
    type: "object",
    properties: {
      trace_id: { type: "string", description: "The trace ID to look up" },
    },
    required: ["trace_id"],
  },
  handler: async (client, args) => {
    const result = await client.getTrace(args.trace_id as string);
    return JSON.stringify(result, null, 2);
  },
  • mcp-server/src/tools.ts:47-53 (schema)
    Input schema for 'get_trace': requires a single string parameter 'trace_id' (the trace ID to look up).
    inputSchema: {
  type: "object",
  properties: {
    trace_id: { type: "string", description: "The trace ID to look up" },
  },
  required: ["trace_id"],
},
  • mcp-server/src/client.ts:49-53 (helper)
    The AgentGuardClient.getTrace() method - makes an authenticated GET request to /api/v1/traces/{traceId} and returns the trace data.
    async getTrace(traceId: string) {
  return this.fetch<{ trace_id: string; events: unknown[] }>(
    `/api/v1/traces/${encodeURIComponent(traceId)}`
  );
}
  • mcp-server/src/index.ts:23-42 (registration)
    The MCP server registration loop: iterates over all tools (including 'get_trace') and registers them with server.tool().
    // Register each tool with the MCP server
for (const tool of tools) {
  const shape = buildToolShape(tool.inputSchema.properties, tool.inputSchema.required ?? []);

  const toolName = tool.name;
  const handler = tool.handler;

  server.tool(toolName, tool.description, shape, async (args) => {
    try {
      const text = await handler(client, args as Record<string, unknown>);
      return { content: [{ type: "text" as const, text }] };
    } catch (err) {
      const message = err instanceof Error ? err.message : String(err);
      return {
        content: [{ type: "text" as const, text: `Error: ${message}` }],
        isError: true,
      };
    }
  });
}

Tool Definition Quality

A3.6/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description must fully convey behavioral traits. It transparently lists what data is included (spans, tool calls, LLM calls, etc.) but does not disclose potential side effects, permissions, rate limits, or error handling, which would improve safety.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is two sentences: first states purpose, second lists contents. No wasted words, front-loaded with key information.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the low complexity (single parameter, no output schema), the description covers the tool's function and return content reasonably well. However, it lacks any mention of error handling or output format, which would make it fully complete.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100% for the only parameter (trace_id), so the baseline is 3. The tool description adds minimal extra meaning beyond 'by its trace ID', as the schema already provides adequate context.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the verb 'Get' and the resource 'full event tree for a specific trace by its trace ID', making the purpose unambiguous. It distinguishes from siblings like query_traces (which likely queries multiple traces) and get_trace_decisions (decisions only).

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

There is no guidance on when to use this tool versus alternatives like query_traces or get_trace_decisions. The description only states what it does, leaving the agent without contextual decision-making support.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/bmdhodl/agent47'

If you have feedback or need assistance with the MCP directory API, please join our Discord server