Prepare an Exact Reference-Image Recreation

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

The description adds behavioral context beyond annotations: it persists state, and notably states that DrawArch does not call another LLM. It also discloses the approval condition. It doesn't mention overwriting or side effects, but no contradiction with annotations.

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?

Three short sentences, each providing distinct information: the action, a pipeline detail, and an approval rule. No unnecessary words, though the second sentence is slightly cryptic.

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 tool with a large nested schema and no annotations, the description does not convey prerequisites, the exact structure needed, or how this step fits into the recreation workflow. It leaves the agent to infer what 'absolute editable scene graph' entails and doesn't mention the plan's required properties.

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 schema has 0% description coverage, but the description does address the `approved` flag by specifying when it should be true. It only vaguely refers to the `plan` as a scene graph, leaving the extensive schema to carry the plan's structure.

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 action ('Persist') and a definite object ('the host vision model's absolute editable scene graph'), distinguishing it from sibling tools that list, validate, or create DrawIO files. The title reinforces this but the description doesn't explicitly contrast with update_reference_plan.

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 implies use for persisting the scene graph before generating the recreation, and it gives a clear approval rule ('Set approved only after...'). However, it doesn't state when to use this tool versus update_reference_plan or create_reference_drawio.

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