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

Annotations already cover safety (readOnly, idempotent, non-destructive), and the description layers on genuinely non-obvious behavior: per-row billing where `limit` caps both rows and cost, an empty result carrying a `_hint` on which filter to relax, and an untrusted-input warning for user-generated content. It also flags ordinarily-invisible classifier noise in `include_face`. This is well beyond what the structured fields convey.

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?

Dense but front-loaded — purpose and routing come first, then return/behavior notes. Nearly every sentence carries operational weight (billing, hint, untrusted content), though the return-field enumeration and the `include_face` aside push it toward the long side for a tool description.

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 22-parameter, no-output-schema search tool, the description supplies the missing return contract (projected columns, default sort, face_* visibility toggle) and the failure-mode hint. An agent has everything needed to call it correctly and interpret an empty result.

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 95% schema coverage the baseline is 3, but the description contributes real meaning the schema lacks: that `limit` governs cost, that `meta_category` stacks with `category` via AND, that non-indexed `sort_by` values scan more rows, and that face filters apply server-side even when `include_face=false`. It does not re-explain the many straightforward filters, which is the right trade-off.

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?

Opens with a specific verb+resource ('Find Instagram accounts by face-analysis demographics') and immediately enumerates the filter families it combines. It explicitly separates itself from get_top_users, search_users, and get_business_users, so an agent can route 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?

States the selection condition directly: 'Use it when you need gender/age/race/emotion; otherwise prefer get_top_users... search_users... or get_business_users'. Both the when and the when-not are explicit, with the reason each alternative exists.

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