Chat with the scene agent

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

With no annotations provided, the description carries the full burden, and it does a good job: it discloses that this is a bounded run, that the server agent may call scene tools mid-turn, that the tool waits by default, and that it returns the event transcript. It also adds the useful instruction that the agent's numbers carry units/scale_source and should be repeated honestly. It stops short of describing side effects or the wait=false behavior, but the disclosed behavior is substantially richer than typical.

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 description is three sentences with no filler. The first sentence frames the core action and scope, the second states the default behavior and return type, and the third gives a practical domain-specific instruction. Every sentence earns its place and the most important information is front-loaded.

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 no annotations and no output schema, the description covers the essential operational context: the bounded run, mid-turn tool calling, waiting behavior, and returned event transcript. The main gaps are edge cases like wait=false behavior, polling semantics, and possible side effects from the agent's mid-turn tool calls, but these are secondary to the core use case and partly inferable from the schema.

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 only 17%, so the description must compensate. It does partially: 'By default waits for the run to finish' explains the wait parameter's default behavior, and 'returns the event transcript' clarifies the output tied to the run. However, it does not explicitly explain poll_s, timeout_s, scan_id, or message, although their names and schema constraints make much of that inferable. The description adds some value but does not fully fill the low-coverage gap.

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 clearly states a specific verb ('chat'), a specific resource ('the scene agent'), and a precise scope: one chat turn, its own bounded run, and a return value of the event transcript. It also distinguishes this tool from siblings like agent_replay or agent_runs by explaining that the server agent may call scene tools mid-turn. This is unambiguous and not a tautology.

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?

The description implies when to use the tool: when you want one interactive chat turn with the scene agent and want the resulting transcript. However, it does not explicitly name alternatives or state when not to use it, such as when you need to replay an existing run or inspect run events. The usage context is clear but the exclusionary guidance is missing.

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