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

Annotations already declare readOnly/idempotent/non-destructive, and the description adds genuinely new behavioural facts: deterministic with no AI inside, nothing logged, confidence flagging for Arabic, and the shape of the response. The 'not an official spelling' warning is exactly the kind of caveat an agent needs to decide whether the output is safe to use downstream.

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?

Front-loaded with the core verb and scope, then caveats, then behaviour. The trailing enumeration of response fields (result, data_confidence, estimated_items[], sources[], data_version, warnings[]) partly duplicates the output schema, which costs some tightness, though the 'deterministic, nothing logged' clause in the same sentence still earns its place.

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 one-parameter, read-only, idempotent tool with full schema coverage, this description covers purpose, inputs, output shape, determinism, privacy, and the main quality caveat. An agent has everything needed to call it and to interpret a low-confidence result.

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?

Schema coverage is 100% for the single 'text' parameter, so the baseline is 3; the description goes slightly beyond by enumerating accepted scripts and warning that certain inputs (Arabic names) yield degraded, low-confidence output. It adds no format or length constraints on the input, so it does not fully compensate for anything the schema lacks.

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?

States a specific verb and resource — 'deterministic romanisation of any script to plain ASCII' — and enumerates the covered scripts (Cyrillic, Greek, CJK, Devanagari, Thai, Arabic). It is clearly distinguishable from siblings like detect_script (identifies script) or normalize_arabic (script-specific normalisation) without opening any schema.

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?

Names concrete use cases — slugs, sorting, matching, and systems that reject non-ASCII — which gives an agent a clear trigger condition. It also flags a caveat case (Arabic names produce a consonant skeleton flagged low confidence). It stops short of explicitly naming alternatives such as detect_script or normalize_arabic as the preferred tools in overlapping situations.

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