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

Annotations already declare it is a mutating (readOnlyHint=false), non-destructive, non-idempotent, open-world call. The description adds genuinely useful context beyond that: 'No money moves here; the user prepays at checkout_url', the blind-judging behavior in verify mode, and the requirement to read the policy resource once. It does not explain what a successful call returns (job identifier, checkout link timing), leaving a modest gap.

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 a crisp thesis ('One verb for the human step'), then mode-prefixed clauses that keep the dense content scannable. It is packed and somewhat long, but almost every clause carries selection-relevant information rather than 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 complex 11-parameter, three-mode tool with no output schema, the description covers mode behavior, trigger conditions, cost flow, and the accompanying policy resource — enough to route and invoke correctly. Remaining gaps are return values and guidance on choosing 'kind', which matter less than mode selection.

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 only 45%, so the description must carry weight — and it does for the critical ones: mode semantics per enum value, answers ('verify: AI answers to judge blind, 1–3; shape_question: optional'), situation ('shape_question: user's words; verify: the exact question'), and job_id implicitly via live mode. It adds little for the remaining params (kind, state, customer_type, ai_model/ai_vendor, agent_id), which stay schema-only.

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 ('summon') and resource (a licensed human), then decomposes the tool into its three modes (shape_question, verify, live) with a one-line definition of each. It also disambiguates from siblings by naming the equivalents ('same as request_review with answers', 'same as join_expert'), so an agent can tell exactly what this does versus ask_human, verify_human, or join_expert.

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 explicit trigger conditions ('when the user asks for certainty, a professional, notarization-grade confidence or what should I even ask') plus autonomous triggers ('stakes are high: doses, deadlines, filings, money, contracts, or your confidence is low'). It also sequences the modes — 'answer only after that' for shape_question — and points to registry://summon-policy for the full policy.

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