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

The annotations declare readOnlyHint: true, but the description explicitly describes a state-mutating behavior: 'Set mark_read:true to flag returned events read.' This directly contradicts the read-only hint, creating an annotation contradiction. Although the description adds valuable context about feed contents and polling, the contradiction undermines trust and earns a score of 1.

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 three sentences long, front-loaded with the verb and resource, and includes no filler. It concisely packs in output format, filtering options, the mark_read side effect, polling suitability, and an alternative endpoint. The extra detail about the evaluator and the direct feed URL is useful and earns it a 4 rather than a 5.

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 tool with 5 optional parameters and no output schema, the description is exceptionally complete. It specifies the return format (source, citation_uri, raw event payload), explains filtering semantics, describes the read-state mutation, and notes polling behavior plus an alternative HTTP endpoint. This gives an AI agent everything it needs to select and invoke the tool correctly.

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?

The input schema already documents all 5 parameters with clear descriptions, giving 100% coverage. The description adds a concrete example for type ("sec_8k") and clarifies the effect of mark_read on subsequent calls, but these are minor enhancements over the schema. With the schema carrying most of the semantic weight, a baseline 3 is appropriate.

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 a specific verb and resource: 'Pull fired events from your subscription feed.' It clearly defines the tool's scope (recent alerts written to a persisted feed) and differentiates it from siblings like list_subscriptions or recent_changes by focusing on fired alerts. The mention of the raw event payload and citation_uri further pinpoints the functionality.

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 provides practical usage context: how to filter by type and since, how to use mark_read for incremental reads, and a note that polling works fine. It also offers an alternative access method (direct GET endpoint) for scripts and dashboards. While it doesn't explicitly state when not to use this tool, the guidance is clear and actionable for most scenarios.

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