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

Annotations declare a non-destructive, closed-world write (readOnlyHint=false, destructiveHint=false, openWorldHint=false), and the description adds genuinely non-derivable behavior: keepAdding turns the dataset into a live feed capped at 1,000 items, one-in-10/one-in-100 sampling, and that the call may return `needs_confirmation` with an approvalId follow-up. It omits any note about what happens to pre-existing dataset contents when adding, which is the one remaining gap for a mutating tool.

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 with the core action and scope, then layers purpose semantics, live-mode behavior, and the next step in a tight sequence with no filler. It is on the denser side for one paragraph, but every clause carries information an agent needs.

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, a 100% documented schema, and no output schema, the description covers the non-obvious flow (new vs existing dataset, live mode caps, confirmation handshake, follow-on tools) without needing to restate return shape. Minor gaps remain around workspace/token handling and interaction with existing dataset contents, both of which are partially covered in the schema itself.

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 baseline is 3, but the description goes further by explaining what each `purpose` value actually retains (review = reply plus tool calls for rating; eval = reply becomes the expected answer; spec = questions only) and the meaning of datasetId vs newDatasetName and the sampling factor. This is real semantic value the bare enum cannot convey.

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 — saving the conversations behind a pick into a new or existing dataset — and immediately scopes it against the sibling narrowing tool caliper_traces_list, which it shares semantics with. An agent can distinguish this from caliper_datasets_add_items and caliper_datasets_create without opening a schema.

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 decision-relevant context: which purpose (review/eval/spec) fits which downstream intent, when to use a new dataset vs an existing datasetId, and explicitly names the follow-on step (caliper_evals_create or a Caliper review). It stops short of stating when not to use it, so it lands just under an explicit when/when-not guidance.

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