delegate

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

Directly discloses the behavior that makes destructiveHint and openWorldHint true: 'Auto-approves every permission the agent requests, in any mode and anywhere on disk.' It also reveals the clarification interaction pattern (prose in result, resume with resumeSessionId) and that invalid values fail before the prompt while naming the accepted set. This adds genuine context far beyond the bare annotations and does not contradict them.

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?

Six sentences, each pulling weight: purpose, prohibition, safety warning, interaction contract, defaults policy, skill pointer. The most critical warning (auto-approve anywhere on disk) is front-loaded in the third sentence, not buried. No filler, no restatement of schema content.

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 9-parameter, 2-required, destructive, open-world tool with no output schema, this is remarkably complete. It covers safety, interaction protocol, defaults policy, per-parameter touch rules, and even the resolution of paths (relative paths resolve against workspace but are not limited to it) and error behavior (invalid effort values fail before the prompt, skips surface in protocolWarnings). Nothing an agent needs to call it 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%, so the baseline is 3. The description earns a 4 by adding a cross-cutting policy the schema alone can't express — the 'touch nothing unless the user asks' rule that applies across fast, model, effort, and context — and by signaling which params are safe to pass only on explicit request. It doesn't need to restate field meanings since the schema already documents them fully.

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?

Opens with a specific verb+resource: 'Delegate a coding task to cursor-agent over ACP.' It goes further and distinguishes itself from the naive alternative with 'Never shell out to cursor-agent — use this tool only,' so the agent cannot mistake its scope. Unambiguous.

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?

Explicit when-to-use and when-not-to-use guidance: the prohibition on shelling out, the warning that it auto-approves every permission in any mode and anywhere on disk, and a clear defaults policy — 'Keep model, fast and effort at their defaults unless the user explicitly asks.' Multiple params get a 'only when the user asks' qualifier, which is exactly the guardrail an agent needs for a destructive, open-world tool.

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