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

Annotations already carry readOnly/idempotent/non-destructive, and the description confirms read-only while adding real behavioral context beyond them: hourly history limited to ~30 days, timestamps in the account time zone, currency for spend/CPC/CPM, conversion lag ('settle over about a day') and the null-vs-zero distinction. Missing only data freshness/pagination limits beyond what limit/offset imply.

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 purpose, then organized clause-by-clause around parameters, with zero filler sentences. It runs long, but nearly every clause encodes a constraint or behavior an agent needs, so the density justifies the length.

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 12-parameter analytics tool with an output schema, the description covers scope, level, granularity, segmentation, zero-impression semantics, conversion caveats, time zone, currency, date handling and paging. Nothing required for a correct invocation is left unstated.

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% with 12 parameters, so the description carries the full load and does so: scope_id prefixes (cmpn_/adgrp_/ad_), level values, granularity values with history caveat, segment enum values (android_app, ios_web, etc.) that the schema lacks, date format and defaults, and paging via limit/offset.

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?

Names a specific artifact (ChatGPT Ads delivery report) and enumerates the dimensions it reports (time buckets, breakdowns, zero-impression rows, conversions), which clearly separates it from siblings like chatgpt_ads_details, chatgpt_ads_status and diagnose_chatgpt_ads_delivery.

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 when to reach for each option, including the diagnostic use case ('include_zero=true ... is how you see what is NOT delivering') and incompatibility rules (segment not with hourly or conversions; conversions only at granularity none/daily). It does not, however, state when to prefer this tool over sibling reporting tools such as diagnose_chatgpt_ads_delivery.

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