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

Annotations already declare destructiveHint=true, openWorldHint=true and non-idempotent, so the safety bar is lower, yet the description still adds real context: the token is short-lived, KYC and DocuSeal signature steps are triggered, and payment is involved. It never warns what a retry does to an in-flight formation, nor that a duplicate call may create a second filing — the one gap against a non-idempotent, destructive operation.

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 action is front-loaded in the first five words, followed by the conditional AI-agent branch and then the two hard gates. The long parenthetical listing cost, timeline, KYC, payment, signature and the not-legal-advice notice is verbose but each item is a genuine prerequisite, so little could be cut without losing meaning.

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 an 8-parameter, nested-object, destructive, open-world creation tool with an output schema, the description covers the gating conditions, owner typing and the downstream KYC/signature flow. It stops short of saying whether the call is asynchronous and that status should be tracked via offshoreproz_get_formation_status, which would close the loop for an agent.

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, but the description adds a cross-parameter rule not stated in the schema: beneficial_owner is always a human custodian even when owner_type='ai_agent', and agent_context.agent_id must accompany that mode. That coupling is the kind of meaning the schema alone leaves to inference.

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?

"Start forming a company" gives a specific verb (start/initiate) and resource (company formation), and the description immediately names the estimate/sibling flow it depends on. It is clear enough to separate from offshoreproz_estimate_cost, but never explicitly distinguishes itself from offshoreproz_retry_formation or offshoreproz_register, which also touch formation state.

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?

Strong on preconditions: it states both hard gates (estimate_token from offshoreproz_estimate_cost and user_confirmed_cost_and_process=true) and enumerates what the user must have seen before that flag may be set. It also gives a conditional branch for AI-agent formations. What is missing is negative guidance — e.g. use retry_formation to resume a failed formation rather than re-creating one.

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