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

Annotations already provide readOnlyHint, openWorldHint, idempotentHint, and destructiveHint. The description goes further by revealing the mark_read side effect that advances the feed, describing the return payload (source, citation_uri, raw event payload), and noting the feed is persisted. This adds meaningful behavioral context beyond what annotations provide, though it doesn't cover auth or rate limits.

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?

Three sentences efficiently deliver the core purpose, return details, filtering, stateful behavior, and polling advice. It is front-loaded with the main verb and resource, and every sentence earns its place without verbosity.

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, the description appropriately summarizes the return payload. It covers filtering, mark_read side effects, and polling suitability. Combined with the 100% schema coverage for parameters, it gives a complete picture of how to use the tool, though it leaves potential error conditions and rate limits undisclosed.

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 100% and each parameter has a clear description. The description adds value by providing a concrete example for type ("sec_8k"), clarifying since as an ISO timestamp, and explaining the practical consequence of mark_read (next call only shows newer events). It doesn't add material meaning for limit or unread_only, but the schema already handles those well.

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 opens with 'Pull fired events from your subscription feed,' which is a specific verb+resource pairing. It then clarifies it returns alerts the evaluator has written, distinguishing it from sibling tools like list_subscriptions (which manages subscriptions) and recent_changes (which tracks other changes). The purpose is unmistakable.

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 indicates when to use it (to retrieve alerts from a subscription feed) and states polling is fine, even suggesting an alternative HTTP endpoint for scripts/dashboards. It does not explicitly name sibling alternatives, but the context clearly implies this is the tool for reading fired events, while siblings like list_subscriptions serve different purposes.

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