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

Annotations already declare the write-idempotent, non-destructive, closed-world profile, so the description's job is to add context, which it does: the job runs in the background, a duplicate start is coalesced into the existing job, and 'Nothing is sent' reassures the agent that no outbound messages result. It does not describe job status/progress reporting, which is the main remaining behavioral gap.

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?

Compact and front-loaded: the action and scope come first, then source values, then preconditions, then retry semantics. The semicolon-heavy packing is dense but every clause adds information, and there is no filler.

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 no output schema, the description covers the important return behavior ('returns that job', 'carries on in the background') and the prerequisites for a mutation-style kickoff tool. It could say more about how to observe progress or what the returned job contains, but an agent has enough to invoke it correctly.

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 only 33% and idempotency_key is the only documented parameter, so the description carries real weight: it supplies the allowed source values (openphone/Quo, gmail, transcripts) that the schema leaves as a free string, and explains days as 'days back' with a default of 90. That is meaningful compensation, but the source list reads as examples rather than a closed enum and the 730 maximum is never surfaced.

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 ('Starts importing the past messages of a connected inbox into Sergio') and anchors it to a known UI surface ('as Import history on Data sources does'). It also enumerates the source values the agent can pick from, which separates it from the nearby spreadsheet-import siblings.

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 concrete preconditions: the account must already be connected, and connecting happens elsewhere (owner, Settings). It also tells the agent what happens on a repeat start ('returns that job') and how to retry (same idempotency_key). It stops short of naming explicit alternatives such as cancel_import or start_spreadsheet_import, so there is no full when-not guidance.

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