Read Session Events

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

Annotations already declare readOnlyHint=true and destructiveHint=false, so the agent knows this is a safe read operation. The description adds valuable context: that events are 'durable' (persisted), pagination is via 'next_seq' and 'has_more', and token-delta telemetry is excluded by default. This goes beyond annotations without contradicting them, providing practical behavioral expectations for the agent.

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 comprised of three concise sentences, each adding distinct value: the first defines the operation, the second clarifies a special parameter, and the third provides pagination guidance. It is front-loaded with the core purpose and has zero fluff. This is exemplary conciseness with high information density.

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?

Given the tool's moderate complexity (5 parameters, pagination) and no output schema, the description covers the essential points: what events are read, the main optional parameter, and how to paginate. It does not explain the output format or the meaning of response_format, but the schema's enum and default handle that. The description is almost complete for effective usage, with only minor gaps in return value documentation.

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 0%, so the description must compensate for the undocumented parameters. The description mentions 'bounded page' (implying limit), 'next_seq' (since_seq parameter), and 'token-delta telemetry' (include_token_deltas). It also states the default exclusion of token deltas, which aligns with the default false. However, it does not describe response_format or the meaning of since_seq beyond pagination, but given the schema has defaults and clear types, this is a solid compensation.

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 states 'Read a bounded page of durable session events' which specifies the verb (Read), resource (session events), and key characteristic (bounded, durable). It distinguishes itself from siblings like session_list (list sessions) and session_metrics (aggregate metrics) by focusing on raw event logs. The mention of token-delta telemetry exclusion further clarifies its scope.

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 strong context: 'Continue with next_seq while has_more is true' gives explicit pagination guidance, and 'Token-delta telemetry is excluded by default; opt in only for streaming forensics' tells when to use the include_token_deltas parameter. However, it does not explicitly mention alternative tools for other event-related operations (e.g., session_metrics for aggregated data), which would strengthen the usage guidance.

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