firestarter_approve

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

Beyond the annotations (destructiveHint, readOnlyHint=false), the description discloses critical behavioral traits: it actually executes a financial transaction and pays, uses the buyer's saved default address automatically, handles price changes via a nonce-based confirmation flow, resolves a single pending purchase automatically, and rejects non-purchasable items. It also details the consequence of no saved address ('approval of physical goods is rejected'). All traits are consistent with the destructiveHint annotation; there is no 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?

The description is long but well-organized: it front-loads the core purpose and then systematically covers lifecycle, address handling, option selection, shipping, price changes, and edge cases. Each paragraph addresses a distinct concern, and every sentence contributes concrete operational detail. While it could be trimmed slightly, the density is justified given the tool's complexity (8 parameters, multiple fallback behaviors). It earns its length.

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?

The description covers nearly every operational scenario an agent might face: missing address, price-change confirmation, auto-resolving single pending orders, multiple pending candidates, informational vs. purchasable options, and shipping-speed selection. It references response elements like PRICE_CHANGED and masked ship-to views, but does not fully specify the successful response structure (e.g., order confirmation details). Given the absence of an output schema, a bit more detail on the exact success response would round it out, but the coverage is already substantial for a tool of this complexity.

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%, but the description adds substantial semantic depth beyond schema entries. For example, it explains that option_id is preferred because it 'identifies the product itself rather than a position that can shift,' warns never to guess confirm_total, stresses consent_nonce is single-use and echoed verbatim, clarifies delivery_address can be a string or object and that it's saved as default on first orders, and details shipping_option_index semantics. This guidance is essential for correct invocation and goes far beyond the schema's already-descriptive parameter definitions.

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 clear, action-oriented statement: 'Confirm and place an order that is awaiting approval — this is the step that actually BUYS and pays.' It explicitly names the resource (an order awaiting approval) and distinguishes the tool from siblings like firestarter_execute (which returns options) and firestarter_receipt/track_order (post-purchase steps). The lifecycle context reinforces the unique purpose.

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 sequences usage: after firestarter_execute or a listing_id buy returns options, this tool places and pays; afterwards use receipt and track. It also gives clear when-not guidance: 'Only Firestarter-purchasable options can be approved — browse-only results are rejected.' It explains when to omit execution_id and how the tool resolves pending purchases, and when to use option_id vs selected_option, plus when to pass delivery_address/address_id. This is comprehensive and leaves little to inference.

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