Update dataset

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

Annotations only say destructive/idempotent=false/readOnly=false; the description goes far beyond by disclosing that prompt/schema/tool edits turn existing cells `stale`, that rename keeps cells while delete drops them from reads, that added input columns read empty, that index_column re-keys every row, and that a 409 `dataset_in_use` occurs during a running fill. It also explains the copy-on-write safety net for shared enrichments, which is exactly the kind of consequence an agent needs before mutating.

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 spec paragraph followed by four examples that each demonstrate a distinct, non-redundant capability. The opening paragraph is very dense — heavy backticking and semicolon chaining make it hard to scan — and the fourth example's `Ticker`/`Website` payload only obliquely demonstrates the stated dedupe 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?

For a 7-parameter tool with deep nested objects, the description covers mutation semantics, budget rules, pinning behavior, and conflict errors thoroughly. It never states what the tool returns on success, which matters slightly more here because no output schema exists to fill that gap.

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?

With 100% schema coverage the baseline is 3, but the description adds semantics the schema only implies: the 1.5-credit-per-column budget drift, `budget_locked` winning over that adjustment, and type inference on added output columns. The four worked examples concretely demonstrate nesting and operation combinations (`{}`, `rename`, `delete`, `tool`, `null` unpin).

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 — 'Edit a dataset's structure (no run)' — and immediately scopes it to name, index_column, input_columns, output, budget. The parenthetical '(no run)' plus 'Then enrich with dataset_id' cleanly separates it from enrich and update_enrichment, whose names alone would otherwise make the boundary ambiguous.

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?

Gives real workflow routing: use this first, then call `enrich` with `dataset_id`, and points to `help` with `datasets`, `pinned_columns`, `output_schema` for deeper reference. 'Edits a shared enrichment on a private copy' hints at when to prefer `update_enrichment` instead, but never names that sibling or states the exclusion explicitly.

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