Diagnose a video and suggest fixes

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

Annotations already declare readOnlyHint=true and openWorldHint=true, so the safety profile is covered. The description adds genuine behavioral context beyond that: the output is conditional on an AI model being connected, in which case a diagnosis, title variants, thumbnail ideas and next actions are appended. It also implies the template parameter drives the voice of suggestions. What is missing is any note on latency or cost of the AI-augmented path, which would matter for an agent deciding whether to leave diagnose=true.

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 performance metrics are front-loaded, and the conditional AI block is correctly placed second so readers learn what they get by default before the extras. The long comma-delimited metric list is dense but every item is a distinct return field rather than filler. Slight trimming of the enumerations would improve scannability without losing information.

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?

There is no output schema, so the description carries the return-value burden and does so thoroughly by listing every reported metric and the optional diagnosis block. Combined with read-only annotations, an agent has enough to call this correctly. The remaining gap is the undocumented video_id format and no indication of what happens when diagnose=true but no AI model is connected.

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 75%, so the bare video_id gap is the main hole and the description does not close it (it never says the ID is a YouTube video ID versus an internal one). It does add meaning for diagnose by tying it to "with an AI model connected it adds a diagnosis," and for template indirectly via "title variants and thumbnail ideas." The channel parameter receives no treatment in the description at all.

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 opens with a specific framing ("Full performance picture for one video") and then enumerates exactly what is returned: outlier score, watch time, average view percentage, hook retention, traffic sources, search terms, subscribers gained and SEO score. That is far more precise than a restatement of the name. It does not, however, name or distinguish itself from nearby siblings such as get_analytics, seo_check, find_outliers or get_video, so an agent must infer the boundary from the output list alone.

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?

Usage is implied rather than stated: the metric list suggests "use this when you want a per-video diagnostic pass." There is no explicit when-to-use, when-not-to-use, or routing advice pointing at alternatives like get_analytics (channel-level) or seo_check over the same video. The mention of AI-dependent extras is contextual information, not guidance about selecting this tool over another.

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