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

With annotations already declaring readOnlyHint, openWorldHint, idempotentHint, and non-destructive behavior, the description adds substantial context: live TNA HEAD check, confidence set to 0.0 on non-200, transient error retry behavior, and the instruction to surface failures rather than fabricate citations. This far exceeds the annotation baseline and provides critical operational details.

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 about 180 words but every sentence delivers essential information: primary use, behavior, failure modes, and critical cautions. It is front-loaded with a prominent uppercase usage directive and structured with clear paragraphs. No filler or redundancy; it earns its length for a tool with network-dependent behavior.

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?

The tool has an output schema covering return values, so the description focuses on behaviors, errors, retries, confidence semantics, and anti-fabrication rules. It covers failure modes (timeout, connection error), retry policy, and user guidance on what to do if resolution fails. Given the complexity of live TNA checks and confidence scores, this is admirably complete without needing further elaboration.

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 single parameter `citation` is fully described in the input schema with explicit examples and length constraints, achieving 100% schema coverage. The description mentions citation types (neutral citation, SI, legislation section, retained EU law) which overlaps with the schema but does not add additional parameter-level syntax or format details. Schema carries the burden, so baseline 3 applies.

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 uses a specific verb ('Parses + resolves') and resource ('a single citation') and immediately distinguishes itself by explaining the use case: 'USE THIS TOOL BEFORE constructing an OSCOLA citation string from known fields, OR to confirm a citation points at a real document.' This clearly differentiates it from siblings like law_citations_parse and law_citations_format_oscola.

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 provides explicit when-to-use guidance ('USE THIS TOOL BEFORE constructing...') and strong when-not-to-use warnings ('Do NOT format or quote a confidence-0.0 citation', 'do NOT manufacture a citation'). However, it does not explicitly name sibling alternatives, only implies that formatting from known fields is a different route. This is clear and actionable but stops short of named alternatives, so a 4 is appropriate.

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