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

Beyond the annotations (readOnlyHint=false, destructiveHint=false), it discloses the critical non-obvious behavior: removing a member ENDS membership as-of a date rather than deleting it, past purchases stay attributed and only later ones stop rolling up. It also explains persistence ('what makes Aurgus remember') and the downstream effect ('the NEXT close matches them automatically'), which is exactly the kind of consequence an agent cannot infer from structured fields.

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 core purpose is front-loaded in the first clause and every sentence carries operational information. It is, however, a single dense paragraph heavy with em-dashes, parentheticals, and ALL-CAPS emphasis, which slows scanning for an agent parsing it against sibling definitions.

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 three-parameter mutation with no output schema, the description covers the input contract, the semantic effect of each path (add vs. end-as-of-date), the immutability guarantee, and the required next step. Nothing an agent needs in order to invoke it correctly and then recover the state 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 100%, so the baseline is 3; the description earns a step up by tying valid_from/valid_to to real-world events (a GPO join, an acquisition, a rename, a member leaving) and by distinguishing the additive 'members' path from the 'ended_members' path in prose. It largely mirrors rather than extends the per-field schema text, so it is not a full 5.

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 ('Record the human's answer to a customer-grouping question') and immediately scopes it to two concrete cases: a grouping compute_close proposed, or a roster change. It explicitly contrasts with the sibling that does the proposing ('grouping compute_close PROPOSED') and with Aurgus's own behavior ('Aurgus never changes a group on its own'), so an agent can tell this apart from compute_close without opening either 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?

Gives an explicit precondition ('Use ONLY after Aurgus surfaced the question and the user answered') and an explicit follow-up ('After confirming, run compute_close again to fold the change in'), which routes the agent through the correct sibling sequence. It also names the input-discipline rule ('pass the agreement_id and the name(s) exactly as they appear in the data') and when to set dates versus omit valid_from.

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