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

The description goes far beyond the readOnlyHint annotation. It explains subtle behaviors: the basis flip (g4↔g3) that changes the headline total, the China-leg publication step that causes a one-time jump, and the provenance tagging (observed vs reconstructed). It also discloses the response structure (presentation object with display_markdown, data.rows, summary), making the tool's behavior fully transparent. No contradiction with 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 extremely verbose, with repetitive explanations about basis flips and China-leg steps that appear multiple times (e.g., 'Reading note, enforced by the artifact' and then again in the same paragraph). While each sentence adds some nuance, the redundancy and long, clause-heavy sentences could be trimmed for clarity. It is structured logically, but not concise.

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?

Given the complexity of the data (vintage handling, basis flips, provenance), the description covers all necessary context: what the data represents, how it's structured in the response (presentation, data.rows, summary), edge cases (China publication step), and defaults. The output schema is described in the prose, so an agent knows exactly what to expect. No gaps remain.

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?

Both parameters (interval, lookback_weeks) have descriptions in the schema, and the tool description adds crucial semantics: defaults, clamping (1–52), and how interval affects the sample cadence. This is high added value beyond the schema, which only lists enums and basic types. The coverage is 100%, and the description enriches both parameters meaningfully.

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 this is a read-only history of global liquidity, returning a timeseries with specific components (Fed/ECB/BoJ/PBoC, basis, classification). It explicitly names the sibling tool get_financial_conditions as its 'companion', distinguishing it as the historical counterpart. The verb 'get' plus the resource 'liquidity_history' leaves no ambiguity about its 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?

It provides strong usage context by naming get_financial_conditions as the current-day counterpart and specifying the default period (last 12 weeks, weekly). It also explains the data provenance (observed vs reconstructed) and the handling of basis flips, which helps an agent decide when to call it. However, it does not explicitly state 'use this when you need historical liquidity data' or contrast with other history tools like get_conditions_history, so a small gap remains.

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