send_to

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

Discloses detailed behavior beyond annotations: it explains what a 'clean verified success' returns, how degraded transport or deduplication adds fields, and that verbose=true yields the full legacy receipt. It also notes that non-success automatically keeps full diagnostics. The annotations (readOnlyHint=false, destructiveHint=false) are not contradicted; the description adds rich behavioral context about receipts and edge cases without conflicting.

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?

While the description is long, it is dense with necessary information and front-loads the primary purpose and key usage rules. Each sentence contributes value: safety constraints, mode behavior, receipt structure, and edge-case exclusions. There is no fluff or tautology; the length is justified by the tool's complexity and 16 parameters.

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 tool with 16 parameters, nested objects, and multiple modes, this description is remarkably complete. It covers target types, mode-specific return fields, failure behavior, the verbose flag, and the collab_path restriction. It also mentions unknown callers and allowed push sources. Nothing critical an agent needs to call this correctly appears 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?

With schema description coverage at only 31%, the description compensates significantly. It clarifies the 'target' parameter by stating targets may be 'one agent, structured agent targeting, or a raw surface in surface/command/key mode,' and explains mode-specific receipts (text/command vs key). It also interprets the 'verbose' parameter by describing the full legacy receipt. This adds substantial meaning to otherwise undocumented parameters.

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 opens with a clear verb+resource: 'Send text or a key through the shared delivery engine.' It immediately establishes the tool's core function and distinguishes it from alternatives by explicitly stating that send_to submits messages rather than the agent sending Return itself, which clarifies its unique role among siblings like wait_for or read_screen.

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?

Provides explicit when-to-use and when-not-to-use guidance: 'Never send a Return yourself for a message; send_to submits messages. Key-Return is for pickers, menus, and permission prompts.' It also gives a concrete alternative for workers with collab_path: 'append to that collab file instead.' These are direct, actionable routing instructions that leave no inference.

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