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

Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, so the safety bar is low. The description still contributes by explicitly scoping 'Previews and suggestions do not save rules or change transactions,' telling the agent which specific paths are side-effect free. It does not mention the odd presence of a 'newName' parameter on a read-only tool, which is a missed opportunity for clarification.

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 appropriately short for a 12-parameter tool and front-loads the three action modes. But the lookup sentence ('Supply name or names to find existing rules by their words, the category or payee they set, or "#N" for the rule at rank N') is dense and hard to parse in one pass, and the trailing 'Set action to one of: list, preview, suggest' should logically lead rather than trail the paragraph.

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?

With no output schema and heavy nested input objects, the description should clarify what list/preview/suggest actually return and how the 'rules' array parameter feeds a preview; neither is addressed. Pagination parameters (before, offset) and 'newName' are left dangling, so an agent cannot confidently form a preview call using proposed rule conditions even though that is advertised as a primary mode.

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 description coverage is only 8% across 12 parameters with deep nesting, so the description must carry weight. It does explain name/names lookup semantics (by rule words, by category/payee set, or '#N' rank), that set and unusedOnly narrow the list, and that when/match define preview conditions. However, 'rules', 'before', 'offset', and 'newName' are entirely unexplained, and the many nested fields under when/set (startsWith, excludes, direction, autoPost, class) get no semantic treatment.

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?

Names a specific resource (automatic transaction categorization rules) and clearly enumerates the three modes of operation: listing rules, previewing matches for proposed conditions, and suggesting rule patterns. The three-way split is not derivable from the tool name alone, so the description adds real discriminative power against siblings like manage_categories or manage_payees.

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?

'Set action to one of: list, preview, suggest' gives explicit mode selection, and 'Use when and match to define preview conditions' gives clear context for the preview path. It stops short of stating when an agent should reach for this tool versus manage_categories/manage_payees/manage_transactions, and never distinguishes preview from suggest in terms of intent.

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