search

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

With no annotations, the description carries the full burden and does well: it discloses scope differences (text hits cover every watch, visual moments only the selected watch_id), that results include a contact sheet of best frames/album images, the score thresholds (strong >= 0.5, possible >= 0.05), and the first-visual-search indexing latency. It stops short of stating read-only semantics explicitly, result limits, or error behavior.

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?

Four dense sentences, each carrying distinct information (capabilities, coverage, scoring, indexing cost), with the core purpose front-loaded. No filler or 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?

For a 3-parameter, no-annotation, no-output-schema search tool, the description covers what is indexed, which watches are covered, what comes back (contact sheet), how to interpret scores, and the indexing delay. The main omission is how the 'visual' flag behaves and any notion of result count or pagination.

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. It adds real meaning for watch_id ('default: the most recent watch', beyond the schema's bare null default) and richly characterizes the query string. But the 'visual' boolean is never named or explained — an agent must infer that visual=false yields text-only results across all watches.

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?

States a specific verb and resource ('Find moments') and enumerates all three search modalities with concrete query examples ('a red car at night', 'a pie chart', 'someone hugging'). An agent can immediately tell this is a multimodal content search rather than a frame retrieval or navigation tool like frame/zoom/look.

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?

Provides useful in-tool context about what kinds of queries work, that all-weak visual scores mean the thing is probably absent, and that negation is unsupported. However, it never routes the agent against the 15 siblings (frame, look, zoom, images, list_watches) — there is no statement of when to prefer this tool over browsing a specific frame or listing watches.

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