agent-trace-auditor

audit_trace

Full audit of an agent execution trace. Detects infinite loops, duplicate calls, tool errors, cost overruns, schema violations, and latency bottlenecks. Returns verdict (pass/review/block), score 0-100, and detailed findings.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "trace": "<trace>"
    },
    "name": "audit_trace"
  }
}

Input Schema

Name	Required	Description
`trace`	Yes	Agent execution steps. Each step: {tool, input, output, model, ts, latency_ms, error}
`budget_usd`	No	Cost cap in USD (default 5.0)
`tool_schemas`	No	Per-tool required field schemas for input validation

Schema Changelog

Changes observed during successful MCP inspections.

Changed4 schema fields changedJun 3, 2026

addedInput schema / properties / budget_usd / description
```
Added value: +"Cost cap in USD (default 5.0)"
```

addedInput schema / properties / tool_schemas

Added value: +{
+  "description": "Per-tool required field schemas for input validation",
+  "type": "object"
+}

addedInput schema / properties / trace / description

Added value: +"Agent execution steps. Each step: {tool, input, output, model, ts, latency_ms, error}"

addedInput schema / properties / trace / items

Added value: +{
+  "properties": {
+    "error": {},
+    "input": {},
+    "latency_ms": {
+      "type": "number"
+    },
+    "model": {
+      "type": "string"
+    },
+    "output": {},
+    "tool": {
+      "type": "string"
+    },
+    "ts": {
+      "type": "string"
+    }
+  },
+  "type": "object"
+}

First observedMay 16, 2026

TDQS

A4/5.0

Behavior3/5

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

With no annotations provided, the description carries the full burden of disclosure. It does add meaningful behavioral detail: it lists detection categories and return values, and it is implied to be a read-only analysis. However, it does not explicitly state whether the tool modifies the trace, how it handles malformed input, or whether there are side effects. The coverage is adequate but leaves several behavioral aspects unspecified.

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: the first states the tool's purpose, the second lists detection categories and the return format. It is concise, front-loaded, and every sentence contributes meaningful information with no redundancy or fluff.

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?

The description covers the tool's purpose, key input parameters indirectly, and return values (verdict, score, findings). Given there is no output schema, the description explains the return envelope well. However, it does not detail the structure of 'detailed findings' or clarify how the required 'trace' parameter is structured beyond the schema, so it is not fully exhaustive for a complex audit tool.

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

Parameters4/5

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

Schema description coverage is 100% with each parameter having a baseline description. The tool description adds further semantic value by connecting budget_usd to 'cost overruns' and tool_schemas to 'schema violations', which clarifies how parameters are used in the audit. This exceeds the schema's generic descriptions, earning a 4.

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 tool performs a 'full audit' of an agent execution trace and enumerates specific issues it detects (infinite loops, duplicate calls, tool errors, cost overruns, schema violations, latency bottlenecks), plus the return values (verdict, score, findings). This is a precise verb+resource with explicit scope, and it distinguishes itself from sibling tools like diff_traces and quick_verdict by being comprehensive rather than comparative or quick.

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

Usage Guidelines3/5

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

The description implies use for deep analysis via 'Full audit', but it does not explicitly state when to prefer this over quick_verdict or diff_traces. No alternatives are named, and there is no when-not-to-use guidance. Usage is inferred from the tool name and description rather than clearly instructed.

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

All agent-trace-auditor tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.