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

The annotations already declare readOnlyHint, idempotentHint and destructiveHint=false, but the description adds behavior those flags do not convey: polling is cursor-based and lossless ('nothing is skipped regardless of poll timing'), and each call has an in-chat rendering side effect comparable to render_surface. That side effect is particularly valuable to disclose next to readOnlyHint, since an agent might otherwise assume polling is purely observational. It stops short of describing failure modes 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?

The definition is front-loaded with the core return contract (strokes plus next cursor) before secondary details like in-chat rendering and the done flag, and the closing sentence gives the termination condition. The first sentence is long and packs three distinct claims, which slightly hurts scannability but every clause carries information.

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?

There is no output schema, and the description fully compensates by naming the return payload (new strokes, next cursor, done flag) and the in-chat rendering behavior. For a two-parameter, cursor-driven polling tool, an agent has everything needed to drive the loop to completion.

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 both parameters already carry descriptions, including that cursor comes from the previous start/poll call and 0 means from the beginning. The prose only restates the cursor chaining already present in the schema, adding no format or edge-case detail beyond it. Baseline 3 applies when the schema does the heavy lifting.

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 states a specific verb and resource relationship: it continues a sketch_agent_start task and returns strokes newer than a cursor. It explicitly distinguishes itself from the snapshot-style alternative by calling itself 'an event log, not a snapshot', and names the predecessor tool sketch_agent_start so the agent knows the pairing without opening schemas.

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?

It gives concrete loop guidance: 'Call again with the returned cursor until done is true', which tells the agent both how and when to invoke it repeatedly. The implicit prerequisite (a prior sketch_agent_start call producing a taskId) is stated. It lacks an explicit when-not-to-use case, such as what to do on error or when the task is already complete, which keeps it below a 5.

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