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

Annotation Contradiction: readOnlyHint=true declares a read-only operation, but the description explicitly states that setting mark_read:true 'flag[s] returned events read so the next call only shows newer ones,' which is a persistent state change. This is a direct contradiction with the readOnlyHint annotation. No other behavioral detail can compensate for this inconsistency.

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 a dense, information-packed paragraph that front-loads the core purpose. Each sentence adds value: return content, filter options, mark_read semantics, and the alternative endpoint. It is slightly long but every clause is purposeful, making it appropriate for the tool's complexity.

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 explains the return shape (source, citation_uri, raw event payload) and covers the side-effect of mark_read. It explains filtering, the external mirror, and pollability. It could add an example output or mention default limit more explicitly, but it is adequately complete for a read-focused tool with well-described 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?

Schema coverage is 100%, so the baseline is 3. The description adds enrichments beyond the schema: it provides a concrete example for type ('sec_8k'), clarifies that since takes an ISO timestamp, and explains the semantic effect of mark_read (affects subsequent calls). These details give the agent a better mental model than the raw schema alone.

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 clearly identifies the tool's function: 'Pull fired events from your subscription feed.' It names the specific resource (subscription feed), the verb (pull/return), and the delivered content (source, citation_uri, raw payload). This distinguishes it from siblings like list_subscriptions (which lists subscriptions) and subscribe/unsubscribe (which manage them).

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 implies its usage via filtering and polling guidance: 'Polls work fine' and the alternative HTTP endpoint for scripts/dashboards. It does not explicitly name sibling tools as alternatives, but the boundary is reasonably clear—this is for reading alerts, not managing subscriptions. Light exclusion guidance is present via the external URL.

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