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

Annotations already declare readOnly and non-destructive, and the schema documents dry_run, so the safety and cost story is covered structurally. The description still adds genuine behavior beyond that: rows are newest-first, each row is dated at the START of its bucket (first_time + i x step_s), null plus 'unavailable' means no data, _pct is percent and _ratio is fraction, and a daily/monthly series is legitimately days or weeks old. Return shape and pagination behavior are not described, keeping this at a 3 rather than higher.

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?

There is almost no filler, but the text is a chain of semicolon-clipped fragments that is dense to the point of being hard to parse, and the FIELDS block partly duplicates the schema's fields enum. The example is useful; the prose would benefit from clearer sentence boundaries and front-loading the primary use case.

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?

With 12 parameters, no output schema and six selectable field groups, this is a complex tool, and the description covers the main axes: series vs all-series mode, window vs history, bucketing and dating conventions, null semantics, unit conventions, and where to find series names and units. Gaps remain around response size/truncation behavior and the computed fields' exact meaning, but nothing essential to calling it correctly is missing.

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%, so every parameter including last/limit/since_ms/until_ms/timeframe/detail already carries its own documentation and the baseline is 3. The description supplements this with cross-parameter semantics the schema cannot express, notably 'No window = current value; since_ms/until_ms/last = history' and that timeframe is macro_intraday-only, but it adds no syntax or format detail for individual parameters.

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 names a concrete verb/resource pairing ('US macro: one FRED `series`... without series, the latest of every series') plus intraday and computed-field variants, so an agent knows it returns macro time series. It is clear about what is returned but written in a compressed telegraphic register that requires careful parsing to extract the scope.

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 gives explicit routing rules: series optional for `macro` but required for `macro_intraday`; `timeframe` applies only to macro_intraday; catalog lookup is delegated to get_system_info. It also steers reading intent ('for a total, an extreme or a change read summary, for the period covered read ranges') and units to describe_fields, though it never states when not to use the tool.

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