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

Beyond the annotations (readOnlyHint, openWorldHint, idempotentHint), the description reveals a non-obvious external trait: 'TikTok search is a bounded, re-ranked pool, not an endless list.' It then translates that into defensive behavior (dedupe on video_id, stop on diminishing returns) — exactly the kind of trait annotations cannot express. No contradiction with the annotations; re-ranking reflects external data volatility covered by openWorldHint, not a side effect.

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?

Two sentences with zero filler. Purpose is front-loaded in the first sentence, and the second packs a dense, actionable heuristic (count, cursor, dedupe key, stop condition) with no wasted words.

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 5-parameter open-world list tool with no output schema, the description supplies the most critical non-obvious knowledge an agent needs: results are finite and re-ranked, so it must page, dedupe on video_id, and know when to stop. It slightly under-serves the no-output-schema case by not sketching the compact result shape, though the dedupe instruction implicitly confirms video_id is returned.

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% — all five parameters carry descriptions, so the schema fully documents q, count, cursor, expand, and sort_type. The description adds only light reinforcement ('ask for count 30, page with cursor') that largely echoes the schema's existing '30 yields the most distinct results per call' and cursor paging notes, so it stays at the high-coverage baseline.

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 'Keyword search over videos' — a specific verb, resource, and modality in one phrase. This cleanly separates it from sibling search tools like tiktok_search_creators, tiktok_search_hashtags, and tiktok_search_lives (different resource types) and from lookup-style tools such as tiktok_video_stats or tiktok_hashtag_resolve.

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 second sentence is a concrete operational recipe — 'ask for count 30, page with cursor, dedupe on video_id, and stop when two pages add nothing new' — which tells the agent how to drive the tool correctly. However, it never names alternatives or says when to prefer this over tiktok_hashtag_videos, tiktok_creator_videos, or tiktok_explore_feed; the 'keyword search' qualifier only implies the use case.

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