WhitePact

Build Governance Incident Record

rai_incident_log

Read-onlyIdempotent

Log AI safety events like PII leaks, jailbreaks, and bias triggers to build an audit trail and feed SIEM systems.

Instructions

Create a structured governance incident record. Used by Security Engineers and AI Risk Analysts to log AI safety events (PII leaks, jailbreak attempts, bias triggers, hallucination incidents) for audit trail and SIEM integration.

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": {
      "incident_type": "<incident_type>",
      "severity": "<severity>",
      "description": "<description>"
    },
    "name": "rai_incident_log"
  }
}

Input Schema

Name	Required	Description
`evidence`	No	Supporting data: prompt, response, scan results, etc.
`provider`	No
`severity`	Yes
`mitigated`	No
`model_name`	No
`description`	Yes	Human-readable incident description
`incident_type`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

First observedv1.2.6Sep 11, 2026

TDQS

C2.7/5.0

Behavior1/5

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

The description explicitly says 'Create' and 'log ... events', i.e. a write/mutation, while annotations declare readOnlyHint=true and idempotentHint=true. This directly contradicts the structured safety metadata. The added context about audit trail and SIEM integration is useful, but the write-vs-read conflict is disqualifying.

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 the core action front-loaded and no filler. Efficient, though it spends words on audience role labels rather than information an agent needs to invoke the tool.

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

Completeness2/5

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

For a 7-parameter, nested-object tool with no output schema and only 29% schema coverage, the description omits severity semantics, evidence schema, and return behavior. Combined with the annotation conflict, an agent lacks enough to call this correctly.

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

Parameters2/5

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

Schema description coverage is only 29% across 7 parameters, so the description should compensate, but it only maps loosely to incident_type by naming event categories. It says nothing about severity levels, mitigated, evidence structure, provider, or model_name, leaving most required and optional fields undocumented.

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+resource ('Create a structured governance incident record') and enumerates the event classes it captures (PII leaks, jailbreak attempts, bias triggers, hallucinations). Clear on its own, but it names no sibling tools, so an agent gets no help distinguishing it from rai_pii_report, rai_compliance, or rai_audit_summary in a 30-tool family.

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?

It gives the intended audience (Security Engineers, AI Risk Analysts) and the purpose (logging AI safety events for audit/SIEM), which implies usage context. However, it never says when to use this instead of the closely related reporting siblings, nor what prerequisites or thresholds trigger a log entry.

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