map

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

readOnlyHint=true already tells the agent this is a safe read, so the bar is lower, and the description still adds genuine context: this is crawled data, blocked buttons surface the dangerous endpoint they would have called, and how requests are attributed to controls. It omits freshness (when the crawl was taken), result size, and whether the map is live or static — the main behavioral unknowns for a read-only crawl.

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 content is front-loaded and every clause carries information; nothing is filler. The opening sentence is a long, dense run-on, but the follow-up sentence cleanly enumerates the parameter modes.

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?

With no output schema and no schema descriptions, the definition does a fair job describing what data exists, but leaves it unclear how results are shaped and sized, how large a full screen dump is, how deep `query` searches, and whether the crawl is current. For a tool whose primary use is deciding whether to click something, that leaves meaningful unknowns.

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 0% with two bare string parameters, so the description carries the full burden — and it does, mapping each parameter to a distinct behavior (absent = screen list, `screen` = full detail, `query` = search across labels, hovers and endpoints). It stops short of saying what form `screen` takes (name vs. id) or what query matching semantics are used, which leaves one ambiguity.

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 states a concrete resource — a crawled map of the native client covering screens, controls, hover text, click effects and fired LCU requests — which is far more specific than the bare name 'map'. It does not, however, distinguish itself from siblings that sound overlapping (snapshot, state, and especially the sibling literally named `query`, which also appears as this tool's parameter name).

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 clearly explains the three invocation modes (no args = list screens, `screen` = one screen in full, `query` = search labels/hovers/endpoints), which is real guidance. But it never says when to reach for this tool versus snapshot, query, css or eval, and never states the intended workflow (e.g. look up a control here before calling click). Usage is implied, not directed.

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