Paper Plane Post

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

Annotations already declare readOnly, idempotent, non-destructive and openWorld, and the description reinforces the safety profile by enumerating the things it will NOT do (create, read contents, charge, mail). What it omits is what the call actually yields for the agent's next step and whether repeated calls re-render or accumulate; the output schema covers some of that, so 4 rather than 5.

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?

Front-loaded with the action and then the constraints; nearly every clause earns its place as a guardrail. The clause 'including after the Claude skill's file-transfer helper' is the weak link — it is allusive and costs the reader parsing effort without a crisp, checkable condition.

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 single-URL, read-only display tool with an output schema present, the description covers purpose, precondition, and forbidden side effects adequately; return-value detail is legitimately delegated to the output schema. Minor gap: it does not say how to recover if the returned URL is stale or already finalized.

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 0% and the single param is an untyped-for-purpose string, so the description must carry the meaning — and it does: the URL must be the exact value returned by start_postcard and must never be fabricated. That is the critical semantic. It stops short of describing the URL's expected lifetime or what happens if the draft was already paid for.

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?

A concrete verb (display the 'Finish your postcard' button) applied to a specific resource (an existing unpaid draft URL). It is explicitly scoped apart from the sibling start_postcard ('Does not create a draft'), so an agent can distinguish the two without opening either schema.

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?

States the precondition (a URL returned by start_postcard), the hard rule ('Only use the exact returned URL; never invent one'), and the exclusions ('cannot look up orders or delivery status'). Both when-to-use and when-not-to-use are explicit.

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

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

Annotations only declare the safety profile (write, open-world, non-idempotent, non-destructive); the description supplies the behavior that actually shapes a call: $7 with postage, no charge or mailing until the user completes on the website, link expires in about a week, optional illustration only if selected on the site, and a conditional challenge flow requiring a proof-of-work answer plus resubmission of all arguments. Nothing disruptive is hidden.

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?

Front-loaded with what the postcard is and what it costs, then rules, then the challenge edge case — a logical order for an agent to read. It is dense but long, and a few instructions (never invent image data) are repeated in the schema, costing some tightness.

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 7-parameter, nested, open-world write tool with an output schema, the description covers the gaps the schema cannot: price, fulfillment timing, link lifetime, what the user sees and does next, and the failure-mode challenge protocol. An agent has enough to decide whether to call and how to recover from a challenge.

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 exceeds it by explaining the routing logic between the three image paths (photo vs photo_base64 for clients without file transfer vs photo_url when already online) and the conditional meaning of is_ai_generated. The challenge parameter's retry contract is reinforced in prose, though the country-code format is restated rather than newly clarified.

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 and resource ("Start a Paper Plane Post postcard") and immediately defines the artifact: a photo printed on a 4×6 card, optional illustrated version, mailed first class for $7, US to 194 countries. It also explicitly disclaims the sibling's territory ("This tool only starts postcards. It cannot retrieve orders or delivery status"), so an agent can separate it from show_postcard without opening a schema.

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?

Gives explicit use and non-use conditions: use photo for uploads/ChatGPT images, photo_url only when the image is already online, "Do not call this tool or invent image data" when no image exists, and do not call it for order lookups. It also resolves the counterintuitive case — if note or recipient is missing, start anyway without asking — and forbids requesting payment details.

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