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

Annotations already declare readOnly/idempotent/non-destructive, yet the description still adds real behavioral context: the 200-row cap, cursor-based pagination via nextCursor, the extra row fields (collaborators, isManual, datesFromSubtasks, url), the includeSubtasks depth semantics, and the projectResolved caveat that an empty list does not mean an empty project. It stops short of describing the full return envelope, but that is a minor gap against a tool whose safety profile is already covered.

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?

Purpose is front-loaded, then filters, then return shape, then routing alternatives and the projectResolved caveat – a sensible order. It is one dense paragraph rather than broken into lines, so the routing and caveat clauses are easy to skim past, but no sentence is 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?

For a 16-parameter, zero-required filtering tool with no output schema, the description supplies pagination, return fields, sentinel values, the parentId/subtask interaction, and the misleading-empty-result warning. Nothing an agent needs to call it correctly or interpret its output is 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 94%, so the baseline is 3, but the description earns a step up by adding semantics the schema only partially carries: the 'me' / 'unassigned' sentinels, the "null"/JSON-null parentId distinction for top-level-only, case-insensitive titleMode, and the cross-parameter rule that includeSubtasks only recurses when parentId is omitted. The customFields operator matrix is left entirely to the schema, which is acceptable given its own long description.

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+resource (list tasks visible to the caller) and immediately enumerates the filter space, so the agent knows exactly what the tool returns. It also explicitly distinguishes itself from get_plan and list_assigned_tasks, which is the sibling-routing signal the agent needs.

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 gives explicit alternatives and the conditions that select them: PREFER get_plan for a whole project tree in one call, PREFER list_assigned_tasks for your own queue, and use list_tasks when drilling parent to children or paginating large projects. It also names a concrete failure mode (projectResolved=false) and prescribes the remedy.

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