build_surrogate

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

Annotations only say non-readOnly, non-idempotent, non-destructive; the description adds substantial context beyond that: background/async execution, survival of client disconnect, held-out fits, and the warning that fewer than 60 solves yields a less reliable error bar. It does not state cost/rate limits for mesh_checks beyond the ~3-solves-per-point note, so not a 5.

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 and information-dense, but the single long run-on sentence mixes workflow, parameter grammar, and caveats without structural separation, and several clauses (e.g., the unit-omission aside) are buried mid-sentence. Every piece of content earns its place, but the structure makes it harder to scan than it needs to be.

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 7 parameters at 0% schema coverage and no output schema, the description covers the costly/risky decisions (n, parameters, outputs, mesh_checks) well and points to surrogate_status for follow-up. The undocumented mode, language, and case_id parameters, plus no word on what a returned handle looks like, leave minor gaps.

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 0%, so the description carries the full burden and largely discharges it: it documents parameter entries as 1–5 objects of path/min/max/unit with a concrete example, the 'any number update_case can change' rule, the CAD/template limitation on varyable dimensions, outputs naming from results metrics with examples and defaults, the n default of 60 and error-bar implication, and mesh_checks costing ~3 solves per point. Gaps remain on case_id, mode (preview/final), and language, which keeps it from being flawless but it is far above baseline.

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+resource ('Build an instant-answer model of this case') and immediately distinguishes the workflow from siblings: build_surrogate trains, surrogate_predict answers, surrogate_status polls. An agent can route correctly from this sentence alone.

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?

Explains the pipeline position (build here, predict later), notes background execution surviving client close, and directs polling to surrogate_status. It lacks an explicit 'when not to use' (e.g., use instant_estimate for a single point), which keeps it below a 5.

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