Create an AI character

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

Goes far beyond the sparse annotations (readOnlyHint=false, openWorldHint=true, destructiveHint=false). Discloses the 2-credit cost, public-by-default privacy with private=true opt-out, background/non-blocking generation with a required polling cadence, the extra cost of make_video, and the return shape (character with first_look_id). No annotation contradiction.

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?

Relatively long, but every sentence carries operational weight: cost, privacy, polling cadence, terminal statuses, and downstream routing. The core purpose is front-loaded in the first two sentences, prompting guidance sits in the middle, and advanced-path details are deferred to the end.

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 10-parameter tool with no output schema and two usage paths, this covers the essentials: return value, async behavior, polling cadence, terminal statuses, cost, privacy, and the downstream dependency on create_clip. Minor gaps: no failure-mode handling (insufficient credits, what to do on 'failed'), and no statement on whether description and dna_json can be combined.

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 baseline is 3; the description adds value by grouping the 10 parameters into two coherent paths (plain-text description preferred vs structured name+age/gender/dna_json) and by giving content guidance for the description field (nationality, age, hair, eyes, face shape, skin, style, realism anchor). It does not clarify edge interactions such as what happens if both description and dna_json are supplied.

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+resource ('Create a new AI character from a plain-text description'), names the input format, and describes the automatic first-look generation, distinguishing it from siblings like generate_image and create_clip. The pipeline context (look id feeds create_clip) makes the tool's role unambiguous.

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?

Provides explicit workflow routing: confirm credits with get_credits, poll list_images every 5-10 seconds, pass the completed look id to create_clip. It also delineates the simple description path from the advanced structured path. However, it never states when not to use this tool (e.g., if only a standalone image is needed, generate_image would be the alternative), so it stops short of a full when/when-not contrast.

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