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

Adds value beyond annotations by detailing the mark_read side effect ('flag returned events read so the next call only shows newer ones') and confirming polling is acceptable. Does not contradict annotations (readOnlyHint, idempotentHint, destructiveHint).

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?

Compact yet informative; front-loads purpose and return fields, then filters, then behavior. Uses a single sentence with semicolons, which is efficient but could be slightly more structured for quick scanning.

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?

Covers return fields, filtering, mark_read behavior, polling suitability, and alternative access method. Missing details on error handling or edge cases but sufficient for an alert-retrieval tool with no output schema.

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?

Enriches schema descriptions with examples (e.g., 'sec_8k' for type, ISO timestamp for since) and clarifies usage of mark_read and unread_only. Schema coverage is 100%, but description provides additional actionable context.

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?

Explicitly states the action ('Pull fired events') and the resource ('from your subscription feed'), with specific return fields (source, citation_uri, raw event payload). Clearly distinguishes from sibling tools like list_subscriptions by focusing on retrieving events rather than managing subscriptions.

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 usage context (filtering by type/since, marking read) and mentions an alternative direct API endpoint for scripts/dashboards. However, does not explicitly contrast with other sibling tools (e.g., when to use this vs. subscribe or list_subscriptions), though the purpose is clear.

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