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

Beyond the annotations (readOnlyHint, idempotentHint, destructiveHint), the description discloses non-obvious behaviors: streaming credentials and playback/replay URLs are deliberately withheld for security, viewing must occur inside the app, and the 'recent' field contains only recent interactions. These are critical operational constraints not covered by annotations, so the description fully carries the behavioral transparency burden.

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 organized into labeled sections (【需要登录】【何时用】【组合链】【口径/坑】), front-loading the most important usage info. Every sentence serves a distinct purpose: usage triggers, alternative routing, default behavior, and critical security caveats. There is no filler or repetition; it is dense but well-structured.

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 getter with no output schema, the description covers the essential operational details: when to call, what it returns (liveId, recent interactions), how to chain with export/start/end tools, and security restrictions. It doesn't enumerate the full return fields, but the hints (liveId, recent) and the context of the chaining tools give an agent enough to proceed. A slightly more explicit return-field list would make it perfect, but it is not a blocking gap.

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 both parameters are documented. The description adds the meaningful default semantics for liveId: if not passed, it returns the latest session. It also clarifies that activityRef accepts either slug or id (already in schema but reinforced). This exceeds the baseline 3 by providing a behavior not evident from the 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 explicitly states the tool's purpose: an organizer checks whether a live session has been created, is currently broadcasting, or has a replay, and can obtain the liveId for exporting interactions. It clearly distinguishes itself from related tools like ensure_activity_live, start_live_session, and end_live_session by naming the specific use cases and the combination chain. The verb-resource pairing (get + live session) is precise.

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 explicit when-to-use scenarios ('主办方问「我这场直播建了没 / 现在在播吗 / 有回放吗」') and routes to alternatives via the combination chain: export_live_messages, ensure_activity_live, start_live_session, and end_live_session. It also notes the default behavior (liveId omitted = latest session) and warns about credential non-delivery, leaving no ambiguity about when to invoke this tool versus siblings.

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