Verify a previously-cited knowledge fact is unchanged. Provide the entry id and the content_hash you stored earlier; returns whether it still matches the current knowledge base. The trust primitive for AI-to-AI fact-checking.
| Name | Required | Description | Default |
|---|---|---|---|
| id | Yes | Stable knowledge entry id | |
| content_hash | Yes | The content_hash from a previous fetch/cite |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare the tool as read-only and idempotent, and the description adds value by explaining the return behavior ('returns whether it still matches the current knowledge base') and framing it as a 'trust primitive.' This contextual information goes beyond the structured annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise and well-structured: a single purpose statement, a usage instruction, and a contextual tagline. Every sentence contributes meaning with no wasted words.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple verification tool with two well-documented parameters and safety annotations, the description covers the essentials: what it does, how to invoke it, and what it returns. The return behavior is described clearly, which is especially important given the absence of an output schema.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema already provides descriptions for both parameters (100% coverage), so the description adds little new semantic value. It reinforces that the content_hash should be from a previous fetch/cite, but does not introduce additional format, constraints, or examples beyond what the schema already documents.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's function with a specific verb ('Verify') and a specific resource ('previously-cited knowledge fact'). It also explains the mechanism (comparing content_hash) and distinguishes itself from sibling tools by its focus on verification rather than search or retrieval.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description clearly implies when to use the tool: when you have an entry id and content_hash from an earlier cite and want to check if the fact is unchanged. It provides a use case ('AI-to-AI fact-checking') but does not explicitly name alternatives or exclusion criteria, which would have made it a 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Most tools have clearly distinct purposes, but the three search tools (search, search_global, batch_search) and the two relation tools (related, isomorph) could cause initial confusion despite clear descriptions. Overall boundaries are clear enough that agents should select correctly with careful reading.
Every tool uses the consistent coldstate_ prefix followed by a descriptive verb_noun pattern (e.g., batch_search, browse_documents, list_indexes). The naming is uniform and predictable, making the toolset easy to navigate.
At 16 tools, the server sits at the upper edge of what feels reasonable, and the calibration guidance flags 16-25 as heavy. However, each tool serves a distinct function in the knowledge-search lifecycle, so the count is defensible but still slightly high for a single server.
The toolset covers the full read-only knowledge workflow: searching, browsing, fetching, citing, verifying, explaining, resolving, and exploring relations. It also includes meta-tools for capabilities, domains, snapshots, and statistics, leaving no obvious gaps for its stated purpose.