schema_explorer_reshard

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

This goes beyond readOnlyHint=true by stating 'nothing is changed' and disclosing that the underlying query 'scans the whole table', plus how sample_rows trades accuracy for cost (random sample scaled up). That is exactly the cost/behavior context annotations cannot carry. It does not describe result shape or how long the scan may take, which keeps it from 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?

The purpose is front-loaded, but the middle sentence is cluttered with backticks and a parenthetical restating the internal GROUP BY query, which an agent does not need in order to call the tool. The sample_rows note is useful but tacked on as a fragment rather than integrated into the parameter story.

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?

No output schema exists, and the description only loosely indicates the return value ('how rows would spread over the partitions ... next to the current skew') without saying what fields or format come back. For a read-only simulation tool, the cost disclosure is the most important thing and is present, but the result contract and the distinction from the script sibling are 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 0%, so the description carries the full burden. It explains two of four parameters well: columns as the candidate shard key and sample_rows as an approximate row count used for a scaled-up random sample. database and table get no explanation at all beyond their names, leaving half the parameters undocumented.

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 uses a specific verb and resource: 'Reshard simulation ... how a table's rows would spread over the partitions with another shard key' and contrasts it with 'the current skew'. That clearly separates it from schema_explorer_skew, but it never mentions the sibling schema_explorer_reshard_script, which an agent could easily confuse with it.

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 implies when the tool is appropriate by noting 'the app runs it only on an explicit click', which warns the agent that this is an expensive, user-initiated action. However, it never states when to prefer this over schema_explorer_skew or schema_explorer_reshard_script, and gives no prerequisites beyond the required database/table/columns.

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