events_read

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

Annotations already declare a safe, idempotent read, so the bar is lower; the description goes beyond them by disclosing the 500-message retention window and truncation reporting, which materially affect interpretation. It does not disclose that the call can block for up to 20 seconds via wait_ms, a meaningful behavioral trait left unstated.

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 compact sentences, each doing distinct work: what it reads and its bound, the filter, then the truncation caveat and alternative routing. Purpose is front-loaded with zero filler.

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 and four params, the description should say what an event records (type, payload, cursor) and how to advance the cursor; it instead only bounds the buffer and mentions truncation. It is workable for a safe read, but an agent cannot predict the response shape or the meaning of wait_ms from anything provided.

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 only 25% (only table_id is documented), so the description must carry the rest. It conceptually covers two params — cursor-based advancement ('with a cursor') and message-type filtering ('Filter by message types') — but gives no accepted type values, cursor semantics, or token format, and leaves wait_ms entirely unexplained despite its 0–20000 range.

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?

States a specific verb+resource ('Read table event history') plus an explicit scope ('up to 500 retained messages') and mechanism ('with a cursor'). It also names the sibling tools that serve the different purpose ('state_read/objects_list are the authoritative current state'), so an agent can distinguish it from those without opening any schema.

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 closing clause routes the agent: use this for history, use state_read/objects_list for current state, and it explicitly flags that this tool's view may be truncated. That is real guidance, but it never says positively when to prefer this tool (e.g., 'to poll for changes since a cursor') and omits any long-poll/wait behavior that would change usage.

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