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

Beyond the annotations it discloses latency (analysis may take several minutes), persistence (the original graphic and quote are saved on the server), commercial limits (indicative quote, excludes delivery, no order placed, no deadline guaranteed), and the human-review flags (manual_review_required / manual_review_available / project_saved). This is materially more than the readOnlyHint=false / openWorldHint=true annotations convey.

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 core purpose and required inputs are front-loaded, then image handling, then limits and the follow-up workflow, which is a sensible order. It is dense and slightly repetitive on the anti-hallucination rules (local path/identifier warning restated as 'don't invent files, URLs, identifiers, base64'), but almost every sentence carries operational weight for a multi-step tool.

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 six-parameter tool with a nested object, an output schema, and a downstream handoff, the description covers the full lifecycle: valid input construction, attachment failure recovery, fallback channel, and post-call routing. With an output schema present, it correctly does not waste space describing return shapes, only the flag semantics the agent must act on.

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 does not express mutual exclusivity among the image sources, and the description supplies that critical rule: exactly one non-empty image source, omit the others, no local paths or bare identifiers in image_file/image_url, and a one-time retry with image_base64 if the attachment fails to arrive. It also gives concrete examples for size and quantity, which compensates for the 67% schema coverage on those fields.

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 verb and resource: it analyzes one project and prices embroidered/sublimated patches. It also names the sibling it hands off to (submit_patch_inquiry), so an agent can distinguish it from get_patch_capabilities and submit_patch_inquiry without reading the 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?

It states exactly when to call it (one design plus size/quantity/backing/real graphic), when to fall back to the fabrykahaftu.com.pl/chatgpt link (only if the real file is truly unavailable), and when and how to escalate to submit_patch_inquiry with purpose=manual_review after project_saved=true. It also says what not to do (do not invent files/URLs/base64, do not re-add the graphic, do not tell the user to click a ChatGPT button).

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