Detect sensitive data in text

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

There are no annotations provided, so the description carries the burden. The description clearly discloses a key behavioral trait: the tool is non-mutating ('without changing it') and reports only the data element types found ('Returns the data element types Strac found'). However, it does not disclose the exact return structure, pagination, or error behavior. At least the non-destructive nature and detection-only scope are stated.

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 a compact three-sentence block that front-loads the core action and scope, then adds the return-value expectation and the usage-dependent sibling alternative. Every sentence contributes useful guidance with no filler or redundancy.

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 detection tool with only two parameters and no output schema provided, the description covers the purpose, the input domain, the output concept, and the key distinction from its sibling. The only minor gap is not clarifying the exact format of the returned element types or how include_matched_text affects the response, but given that the tool itself is conceptually simple and the return value is summarized, this is adequate.

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 describes 'text' only as a generic string, but the description clarifies the domain and purpose of that text, namely that it contains potentially sensitive data and the tool returns the types found. For include_matched_text, there is no schema description at all, and the main description does not explicitly explain it either, though the boolean default is self-explanatory. Overall, the description adds meaningful semantic context beyond the bare schema.

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 begins with a clear verb-object pair, 'Detect sensitive data in text without changing it', and enumerates the exact categories (PII, PHI, PCI, credentials) the tool detects. It also explicitly contrasts the tool with redact_text, which distinguishes it from the sibling tool. The title alone would not convey this scope, but the description fully specifies the tool's unique function.

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 states both when to use this tool — 'to decide whether text is safe to send onward' — and when to use the alternative, redact_text, 'when you need the sanitised text itself.' This explicit when/when-not guidance with a named sibling alternative is exactly what an agent needs for correct selection.

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