Skip to main content
Glama
GodPrompt MCP Server

GodPrompt MCP Server

Audit reported agent trajectory

audit_agent_trajectory
Read-onlyIdempotent

Reviews bounded, sanitized agent event metadata to detect repeated failures, scope drift, stale verification, and premature completion for read-only trajectory audits without raw logs.

Instructions

Read-only review of bounded, sanitized event metadata for repeated failures, scope drift, stale verification and premature completion. No raw logs or independent execution evidence; cannot authorize an agent.

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": {
      "events": "<events>"
    },
    "name": "audit_agent_trajectory"
  }
}

Input Schema

NameRequiredDescriptionDefault
eventsYes
budget_tokensNo

Output Schema

NameRequiredDescriptionDefault
statusYes
warningYes
findingsYes
truncatedYes
token_budgetYes
events_reviewedYes
reported_tokensYes
reported_writesYes
reported_completionsYes
completion_authorizedYes
independently_verifiedYes
reported_successful_verificationsYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv1.4.2

TDQS

A3.7/5.0
Behavior4/5

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

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so the safety profile is covered. The description goes beyond that by disclosing the evidentiary limits the annotations cannot express: input is bounded and sanitized, no raw logs or independent execution evidence exist, and the output cannot authorize an agent. It stops short of stating what happens when the input exceeds the bounds (maxItems 150, budget_tokens).

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

Conciseness4/5

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

Two tight sentences with no filler, and the positive capability is front-loaded before the limitation clause. The density is high but readable; nothing is repeated from the schema or annotations.

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?

An output schema exists, so return values need no explanation, and the annotations cover the safety profile. For a two-parameter read-only audit tool the description supplies the essential framing (what evidence exists, what conclusions are off-limits), with only minor gaps around budget_tokens behavior and whether the events array must be chronologically ordered.

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 0%, so the schema carries no prose semantics at all, yet the description never mentions the 'events' array or 'budget_tokens' by name. It does implicitly map its four analysis targets onto schema enums (scope drift to scope='outside', stale verification to kind='verification', premature completion to kind='completion'), which is genuine added meaning, but 'budget_tokens' and the ordering/id requirements of the events array are left entirely unexplained.

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

Purpose4/5

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

States a specific verb ('review') and a specific resource ('bounded, sanitized event metadata') and enumerates the four defect classes it looks for: repeated failures, scope drift, stale verification, premature completion. An agent can distinguish this from audit_agent_instructions or review_work_evidence because the object is the agent's own event trail, though no sibling is named explicitly.

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?

Usage is implied by the enumerated signals being searched for, and the second sentence usefully defines the negative boundary ('cannot authorize an agent'), which tells the agent not to route authorization decisions here. However, there is no explicit when-to-use statement and no comparison to the nearest alternatives such as review_work_evidence or diagnose_failure.

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