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

Annotations only declare readOnlyHint=false/destructiveHint=false, so the description carries real weight here and delivers: nothing is sent before the user acts, a session ID is issued, the per-channel shape of the response differs, and auth is required ('needs a Proof account', how to authenticate). It also discloses that clients fold tool results away, a behavioral detail not derivable from any structured field.

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 core flow, then an 'Agent usage' block and a client-rendering caveat — all substantive. It is long for a one-parameter tool and slightly over-explains channel output variants, but every sentence carries actionable instruction, so little is wasted.

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?

With no output schema, the description fully compensates by enumerating return fields per channel, and it also covers prerequisites, the next tool to call, and the rendering constraint ('never hand a qr_text to a link renderer'). Nothing an agent needs to call and follow up correctly 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% and the enum self-documents, so the baseline is 3; the description earns extra credit by explaining what each channel value actually implies at runtime (deep_link/qr_code for telegram+whatsapp, sms_message/sms_dids for sms). That is genuine semantics beyond the enum labels, though it focuses on outputs rather than the argument itself.

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 ('Start adding a new phone number') plus the distinctive mechanism ('by reverse OTP'), which clearly separates it from siblings like start_add_email, create_verification, or start_2fa. An agent can identify the tool's job without opening the 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?

Explicitly describes the flow ('the user starts the exchange themselves and nothing is sent before they do'), names the downstream step ('then poll get_add_phone_status'), and routes per-channel usage (pass deep_link to render_auth_link on telegram/whatsapp; show sms_message/sms_dids on sms). Alternatives and exclusions are stated rather than implied.

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