| Name | Required | Description | Default |
|---|---|---|---|
| method | Yes | Settlement method | |
| period_end | No | Period end (ISO 8601) | |
| media_buy_id | No | Settle specific media buy | |
| period_start | No | Period start (ISO 8601) | |
| publisher_id | No | Settle all for a publisher |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already indicate a non-readonly, non-destructive operation. Description adds behavioral context: Stripe/x402 are only stubs, and it describes the output structure (gross, fees, net) despite no output schema. No contradictions 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?
Well-structured with four short sections. The main description is a single sentence. No filler; the stub warning appears twice but is acceptable for emphasis.
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?
Given the tool has no output schema, the description compensates with an output_format section. It also covers use cases and parameter guidance. However, it stops short of explaining edge cases like what happens when no pending payments exist or whether settlement is a one-time irreversible action (though annotations hint at idempotency=false).
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 covers all five parameters with descriptions (100% coverage). Description adds value by explicitly stating method='manual' works for CSV and Stripe/x402 are stubs, clarifying the enum semantics. Combination hints reinforce filter usage but don't add syntax beyond schema.
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 immediately states 'Settle pending payments for media buys,' which uses a specific verb and resource. It further differentiates from sibling tools by listing settlement methods, clearly positioning it as post-campaign revenue collection rather than creation or reporting.
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 when-to-use scenarios ('publisher wants to collect earned revenue' / 'advertiser needs to settle outstanding charges') and method-specific guidance ('Use method='manual' for CSV export'). It also warns that Stripe and x402 are stubs, effectively telling the agent not to use them yet. Suggests a combination with get_campaign_report.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Most tools have distinct purposes, but there is some overlap between get_product and nexbid_product, which are explicitly described as aliases, and between list_products and nexbid_search with content_type='product'. This could cause confusion, though descriptions help clarify. Other tools like activate, pause, and cancel are well-differentiated for media buy lifecycle management.
The naming is mixed with no consistent pattern. Some tools use verb_noun (e.g., create_media_buy, list_inventory), others use noun_verb (e.g., nexbid_search, nexbid_purchase), and some are single verbs (e.g., activate, pause, cancel). While readable, the lack of a uniform convention across the set reduces predictability.
With 19 tools, the count is on the higher side but reasonable for the dual domains of media buying and marketplace discovery. It covers operations like listing, creating, managing, and reporting, which justifies the number. However, it borders on feeling heavy, especially with overlapping tools like get_product and nexbid_product.
The tool set provides comprehensive coverage for both media buying (create, submit, activate, pause, cancel, track, settle, report, compliance) and marketplace discovery (search, categories, product details, purchase, order status). There are no obvious gaps; workflows are well-supported with clear combination hints, ensuring agents can handle end-to-end tasks without dead ends.