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

Annotations (readOnlyHint, openWorldHint) only tell the agent this is a safe open read; the description adds substantial behavioral context beyond that: it returns the platform's track rather than modeling audio, returns empty_result (free, never charged) on no data, is shipped as experimental with a stable shape, and costs 1 credit. This is exactly the kind of disclosure annotations cannot 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-loads the key warning and every sentence contributes (cost, empty behavior, experimental status). It is slightly verbose in the model-vs-platform framing, but no sentence is wasted.

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?

No output schema exists, yet the description explains the return shape (one text block plus timed cues), the empty case, and the credit cost. For a read-only tool with no output schema, nothing an agent needs to call it correctly is missing.

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?

At 50% schema coverage, url is documented in the schema ('Reel, post or IGTV URL, or a bare shortcode') but cache_max_age is undocumented in both schema and description. The description adds no parameter-level syntax or format meaning, so it neither compensates for the gap nor degrades below the 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?

States a specific verb+resource (caption track published by Instagram for a reel/video) and clearly demarcates it from model-based transcription ('never transcribe the audio ourselves'). An agent knows exactly what this returns and that it is Instagram-specific, distinguishing it from the many *_transcript siblings.

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 'Read this before relying on it' passage is unusually strong when-to-use guidance: it warns that 12 real payloads carried no caption track, so the tool returns empty_result. It stops short of naming an alternative transcript tool to use instead, but the practical usage context is explicit.

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