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

Beyond readOnlyHint=true, it discloses login requirement, the permission model (public readable when logged in, private requires membership/invitation), a 404 pitfall with a named non-workaround, exclusion of hidden and mutually-blocked messages, and pagination continuity via reachingLimit/nextSinceOffsetMs. It even flags a same-millisecond boundary that can drop items — substantial behavioral context the annotations do not carry.

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?

Front-loaded with login requirement and the core export action, then permissions, then pagination, then the posting alternative — a sensible order. The telegraphic semicolon-separated style packs several distinct facts per line, which is dense but mostly earns its space; only the repeated 1000/2000 defaults are redundant with the schema.

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 compensates by describing response-adjacent behavior: reachingLimit, nextSinceOffsetMs continuation, the 1000/2000 bounds, and which messages are omitted. Combined with the access-model and 404 notes, an agent has enough to call this correctly against a 4-parameter read tool.

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 schema already documents kind, limit, liveId, and sinceOffsetMs; the description repeats the 1000/2000 defaults rather than adding to them. It does add semantics beyond the schema by naming the response-side continuation field nextSinceOffsetMs and the same-millisecond gap, though its 'offsetMs' phrasing does not exactly match the schema's 'sinceOffsetMs'.

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?

Opens with a specific verb plus resource ('导出...互动时间轴') and immediately scopes it to '本人有受众权限的' and paginates '按offsetMs'. It distinguishes the two modes ('kind=QUESTION取提问') and names the sibling used for the opposite operation ('发言用post_live_message'), so an agent can separate it from post_live_message without opening the 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?

Explicit routing is given for prerequisites: get_activity or get_organizer_live to obtain liveId, and post_live_message when the goal is to post rather than export. It also states a negative condition — a 404 cannot be worked around with join_activity_online — which is exactly the kind of when-not guidance that prevents wrong sibling calls.

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