End Session

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

With no annotations, the description carries the full burden and does so well: it discloses that the server keeps listening and the next call starts fresh, enumerates the accepted outcome values, and spells out the validation gates (screen read required after certain goal types, findings must meet the plan's target_count). These are exactly the behavioral traits an agent could not infer from the schema.

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 most important and surprising fact — that the server keeps listening — is front-loaded in the first clause, followed by parameter semantics. Content is dense and mostly earns its place, though the goal_type enumeration list and the refusal rules make it a fairly long block of prose rather than tightly chunked guidance.

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?

An output schema exists, so return values need not be explained, and the description covers the refusal semantics an agent must anticipate. The remaining gap is that 'device' and the reason-versus-summary distinction are never explained, which matters for a tool that can be called against a specific target device.

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 description coverage is 0%, so the description must compensate, and it does for the three semantically rich parameters: reason (short summary), outcome (success/partial/failure with definitions), and goal_type (an explicit enum list absent from the schema). It does not explain the 'device' parameter or the difference between 'summary' and 'reason', leaving two of five parameters undocumented.

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 precise verb+resource statement ('Mark the end of the task') and immediately clarifies the non-obvious scoping behavior ('the server keeps listening; the next call starts fresh'), which prevents an agent from assuming the session terminates. No sibling tool competes for this purpose, so differentiation is inherent. The definition is unambiguous about what the tool does.

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?

It gives explicit invocation rules — when success is refused (no screen read after send_message/send_email/purchase/post; fewer findings than the plan's target_count) and that failure is never refused. That is strong, actionable conditional guidance. It stops short of stating when not to call end_session at all (e.g., before completing the plan) or naming alternatives, so it is not a full 5.

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