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

Beyond the readOnlyHint and destructiveHint annotations, the description discloses important behavioral traits: the match_count cap of 200 and saturation on broad/hot topics, that 'similar' uses a default 0.70 cosine floor, and that the 'cites' relation uses a 30-day window. It even warns how to tune around these behaviors rather than relying on match_count alone.

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 dense but every sentence earns its place: purpose, return contract, ranking defaults, anti-noise tuning, the similarity cap caveat, and authentication/read-only status. The most important framing ('Dry-run ... WITHOUT creating a watch') is front-loaded, and the warning is clearly separated with 'NOTE.'

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 complex tool with nested criteria and multiple filter modes, the description is remarkably complete. It covers when to use it, how to iterate, what values to expect, what cap behavior to watch for, and how it relates to create_watch. The presence of an output schema and full schema coverage lowers the burden, and the description still exceeds it.

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?

Although the schema already has 100% parameter coverage, the description adds substantial interpretive value: it explains that 'rising' is the default ranking, that min_impact_pct creates a 'what-is-rising-now' watch, and that match_count saturates at 200 for similar filters. This goes beyond the raw schema and helps the agent choose and tune parameters intelligently.

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 specific verb-resource pair: 'Dry-run a structured filter over recent papers WITHOUT creating a watch.' It clearly frames the tool as the tuning loop preceding create_watch, and the returned fields (window_days, needs_similarity, match_count, sample) make the purpose concrete and distinct from sibling tools like check_watches or create_watch.

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 explains when to use this tool: iterate on a filter 'before saving with create_watch.' It also gives operational guidance for tuning, such as raising min_novelty, switching the collection relation, and using min_impact_pct for an anti-noise watch. The alternative create_watch is named directly, giving the agent a clear decision boundary.

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