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

The annotations already flag destructiveHint=true, and the description adds concrete consequences: visual assets reset to not_started, rendered assets will need regeneration, while continuation links, SFX, and overlays are preserved. It also clarifies that an LLM performs the rewrite, which informs expectations about output variability and the need for a dry_run preview. This goes well beyond the annotation's bare destructive flag.

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 three sentences with no filler: the main purpose comes first, followed by key side effects/next steps, then the alternative tool. Every sentence earns its place and the formatting with em-dashes and parentheticals keeps it readable.

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 destructive mutation tool with no output schema, the description covers the essential context: what the tool does, what is preserved, what is reset, what follow-up action is required, and which sibling to choose when appropriate. The parameter semantics are fully handled by the schema and the behavioral context is thoroughly disclosed.

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?

The input schema already documents all four parameters with high coverage, so the baseline is 3. The description adds extra meaning beyond the schema by explaining what happens to user_input (an LLM rewrites prompts rather than writing verbatim) and by framing the dry_run consequences in terms of asset resets and re-rendering. This is modest but real added semantic value.

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 uses a specific verb ('Rewrite') and clearly identifies the resource ('one segment's creative direction') and the mechanism (an LLM rewrites the shot's prompts). It also distinguishes itself from the sibling update_segment_prompts by contrasting natural-language feedback with exact prompt text, so an agent can select the right tool without opening schemas.

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 description explicitly states when to use this tool: when the user provides natural-language feedback about a shot. It also names the alternative update_segment_prompts and tells the agent to use that instead when exact prompt text is already known. It additionally instructs the agent to re-render assets afterward via generate_segments or regenerate_segment_asset, covering the post-invocation workflow.

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