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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive), so the bar is lower, and the description still adds real behavioral value: it describes dependency propagation ('dependencies of a must become must'), packing order, the returned payload, and concrete failure modes ('rejects duplicate ids, unknown dependencies and cycles with the chain named'). That failure disclosure is genuinely useful and not obtainable from the annotations.

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?

The verb and output are front-loaded, and every clause carries information (rules, propagation, packing, returns, rejection cases) rather than filler. The trade-off is density: it reads as one long comma-spliced sentence, which slightly hurts scanability but not content.

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 complex tool with a nested feature array, 5 parameters, and no output schema, the description is notably complete. It enumerates the return contents (tier, rule, reason, effort totals vs capacity, specific fitting cuts) and the error contract, leaving no obvious gap for an agent to call 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 description coverage is 80%, so the baseline is 3, but the description adds meaning beyond the schema by explaining how parameters interact: must-have propagation through depends_on, should-haves ordered by 'value per effort point in dependency order', and effort totals measured against capacity. That interpretive layer is not present in the schema.

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?

The description names a specific verb and resource ('Sort a feature list into must, should and not now') and the exact rule base (15 rules on tags, manual alternative, buyer demand). It is unambiguous what the tool produces. What it lacks is any explicit differentiation from siblings like build_scope_and_sequence or sequence_build, which an agent might reasonably confuse with a prioritization tool.

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?

Usage is implied by the input (you have a feature list and a build window, and want an MVP cut), and the described algorithm makes the scenario clear. However, there is no explicit when-to-use statement, no when-not-to-use, and no routing to the alternates among siblings such as build_scope_and_sequence.

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