refresh_cache

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

Adds real behavioral context beyond the annotations: the refresh is forceful (re-download, not validate), and 'CVM rewrites the current period in place' discloses a side effect. However, it says nothing about cost/latency of a full re-download, whether the operation blocks, or what happens to partitions not covered by the period filter. It also sits in mild tension with readOnlyHint=true, since a cache rebuild does mutate local state — an agent gets no help resolving that.

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?

Two short sentences, front-loaded with the action and followed by the decision cue — no filler. The one blemish is the unexpanded 'CVM', a domain acronym that carries no meaning for an agent unfamiliar with the data source.

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 3 parameters at 100% schema coverage, an output schema, and annotations present, the description only needs to convey purpose and timing — both are covered. The remaining gap is operational: no sense of cost, duration, or blast radius when period is omitted and everything cached is refreshed.

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 description coverage is 100%, so the schema already defines table, period (YYYY/YYYYMM) and dataset semantics; the baseline is 3. The description modestly adds meaning by tying 'period' to staleness ('the latest month looks stale') and indicating that the operation is destructive of existing cached content before rebuilding. It does not clarify refresh scoping interactions between table and period.

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: 'Force a re-download and rebuild of cached partitions.' An agent can immediately tell this is a cache-maintenance action rather than a data read. It does not name or contrast with the adjacent siblings (cache_status, load_dataset, run_query), so the distinction from those is left implicit.

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?

'Use this when the latest month looks stale' gives a concrete triggering condition tied to the period parameter, which is more than most definitions offer. It stops short of explicit exclusions — e.g., don't use it for a full reload when only one table is stale, or when cache_status already shows the partition is current.

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