Notes

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

The description goes well beyond the annotations. It discloses that the tool decrypts data locally, that the data is sensitive free text, that it is never re-exported, and that it carries a coverage block with specific fields that must be interpreted in particular ways. It also warns against reporting a limit-shaped window as the extent of the data, which is a behavioral trap the agent would otherwise fall into. The annotations (readOnlyHint, idempotentHint, destructiveHint) are consistent with the description, and the description adds substantial behavioral context.

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 dense and information-rich, with the most important facts front-loaded: what the tool does, that it is sensitive, and the key fields. The coverage block explanation is lengthy but necessary because it prevents a specific misinterpretation. It earns its length, though it could be slightly tightened by moving the 'SENSITIVE free text' warning after the first sentence without losing clarity.

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's complexity (sensitive data, coverage semantics, filtering, and a large sibling list), the description is remarkably complete. It explains the return value's coverage block, how to interpret it, how to filter, and what not to do. The output schema exists, so the description doesn't need to enumerate return fields, but it explains the non-obvious parts of the output. An agent has everything it needs to call this tool correctly and interpret its results.

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 input schema has 0% description coverage, so the description must compensate. It explains target_kind by listing the kinds ('sleep' | 'menstrual' | 'mood' | 'general') and how they are authored, which adds meaning beyond the schema's bare anyOf string/null. It also explains limit in the context of coverage.more_available and first_day, giving it semantic weight. The fresh parameter is not explicitly explained, but the description's focus on decryption and local processing implies its purpose. This is strong compensation for a schema with no descriptions.

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: 'Decrypt recent free-text notes (day annotations) locally.' It clearly distinguishes this from sibling tools by explaining that notes are free-text day annotations, that they are decrypted locally, and that they are sensitive. It also names the kinds of notes and how they are authored, which differentiates it from log_note and other get_* tools.

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 tells the agent when to use this tool and how to use it: pass target_kind to filter, join against same-day metric data for pattern analysis, and re-read with a larger limit when coverage.more_available is true. It also gives clear guidance on how to interpret coverage fields, which is essential for correct usage. It does not explicitly name alternatives, but the sibling list is large and the description's specificity about notes vs metrics makes the usage context clear.

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