ai_understand_screen

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

With no annotations, the description carries the full burden of behavioral disclosure. It does disclose that the tool performs a server-side LLM call and requires specific environment configuration, which is a meaningful behavioral constraint. However, it does not mention what happens if the key is missing, whether the screen data is sent externally, or any latency/cost/privacy implications of invoking a remote LLM.

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 two-sentence description is tight and efficient: the first sentence states the core purpose and expected output, and the second sentence adds the critical configuration requirement and fallback. No filler or redundant restatement of the tool name.

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?

Given the zero-parameter input and the presence of an output schema to explain return values, the description covers the essential invocation context: purpose, configuration prerequisite, and alternative path. It could be slightly more complete by addressing what the tool returns or what conditions make it inappropriate beyond missing configuration, but overall it is sufficient for an agent to decide and call correctly.

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?

The tool has zero parameters, so there is no parameter semantics burden on the description. The baseline for zero-parameter tools is 4, and the description appropriately focuses on the tool's behavior and prerequisites rather than 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 clearly states the tool's specific action: ask the server-side LLM to semantically analyze the current screen, including output categories like page type, purpose, and key elements. This verb+resource+result framing distinguishes it from sibling tools like get_screen_elements, which retrieves structural elements rather than semantic analysis.

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 explicitly states when to use this tool: when OPENAI_API_KEY or equivalent LLM configuration is available. It also provides a clear alternative and fallback path: otherwise prefer get_screen_elements + screenshot and analyze yourself. This is strong usage guidance with a named alternative and condition.

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