Skip to main content
Glama

audit_knowledge_base

Read-onlyIdempotent

Audit support documentation offline before RAG indexing to detect conflicting policies, unqualified commitments, missing headings or escalation contacts, and prompt-injection patterns.

Instructions

Deterministically audit support documentation before indexing it into a RAG or AI customer agent. Runs completely offline with zero network calls or API key required, checking one or more Markdown, plain-text, or JSON files for conflicting policy time windows, unqualified commitments, currency ambiguity, missing headings, missing human escalation contacts, and embedded prompt-injection patterns. Returns JSON with score (0-100), grade (A-F), findings[], extractedFacts, recommendations[], and mode ('local_deterministic').

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "audit_knowledge_base"
  }
}

Input Schema

NameRequiredDescriptionDefault
titleNoHuman-readable document title used in audit findings when supplying inline content.Inline Knowledge Base
contentNoInline Markdown, plain-text, or JSON policy documentation to audit. Provide content, filePath, or documents.
filePathNoLocal filesystem path to a readable UTF-8 Markdown, text, or JSON file to audit.
documentsNoCollection of multiple policy documents to audit together so cross-document contradictions (e.g., 14-day vs 30-day refunds) are detected.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv1.2.1

TDQS

A4.4/5.0
Behavior5/5

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

Annotations already declare readOnly/idempotent/non-destructive/closed-world, and the description adds genuinely new behavior: it runs completely offline with zero network calls and no API key, and it is deterministic ('local_deterministic' mode). It also discloses the return payload shape (score, grade, findings, extractedFacts, recommendations), which matters because there is no output schema.

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?

The purpose and pre-indexing context are front-loaded, and both sentences carry real information. The defect enumeration and return-field list are long, but since there is no output schema the enumeration is load-bearing rather than filler; it reads as a dense run-on but no sentence is wasted.

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

Completeness5/5

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

For a read-only audit tool with four fully described optional parameters and no output schema, the definition supplies everything needed: what it checks, that it works offline, what the inputs may be, and what the response contains. No meaningful gap remains for an agent to call it correctly.

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?

All four parameters are fully documented in the schema, including the 'provide content, filePath, or documents' selection rule, so the schema is doing the heavy lifting. The description echoes the accepted formats and the multi-document case for cross-document contradiction detection but adds no syntax or precedence detail beyond the schema. Baseline 3 applies at 100% coverage.

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?

States a specific verb and resource ('audit support documentation') and then enumerates the concrete defect classes it detects (conflicting policy time windows, unqualified commitments, currency ambiguity, missing headings, missing escalation contacts, prompt-injection patterns). Nothing in the sibling list overlaps with this function, so an agent can pick it without ambiguity.

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

Usage Guidelines4/5

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

Gives a clear usage context — run it 'before indexing it into a RAG or AI customer agent' — which tells the agent where this fits in a workflow. It does not name exclusions or an alternative tool to use instead when the input is, say, a live support conversation rather than a document, so it stops short of full when/when-not guidance.

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