notarize_inference

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

Annotations provide readOnlyHint=false and destructiveHint=false, and the description complements this with critical behavior: the notary does not store prompt/response (privacy), costs $0.001, and requires specific private keys. It also details the return payload. This goes beyond annotations, though it does not cover failure modes or idempotency nuances. 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?

The description is compact and well-structured. The main action is front-loaded in the first sentence, followed by essential details (return type, privacy, cost, key requirements). Every sentence adds necessary information without redundancy. It is appropriately sized for a tool with payment and privacy implications.

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 paid, side-effecting tool with no output schema, the description provides a comprehensive picture: it lists return items (attestation, hash, Merkle status), privacy guarantees, cost, and required credentials. It covers the prerequisites and expected behavior thoroughly, leaving little ambiguity for an agent deciding to call it.

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 has 100% parameter description coverage, so the baseline is 3. The description adds value by specifying that {prompt, response, model_id} are the triple used for hashing and that client_timestamp is optional and included in the content hash. This clarifies the relationship between parameters and the receipt's integrity, elevating the score above baseline.

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: 'Get a cryptographic receipt for one AI inference.' It clearly states what is returned (signed Ed25519 attestation, sha256 hash, Merkle status) and what is not stored, distinguishing it from batch notarization and verification siblings. The purpose is unambiguous.

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 says 'one AI inference,' implying not for batch use, and mentions the payment requirement (x402 with wallet keys), which is a critical prerequisite. However, it does not explicitly name alternatives like notarize_batch or get_receipt, leaving some routing to the agent. The context is clear enough but lacks direct when-not-to-use guidance.

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