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

Adds real behavioral context beyond the annotations: matched titles update existing memories rather than creating duplicates, and the 500-item cap. Since annotations already flag destructiveHint=true and idempotentHint=false, this disclosure of the overwrite rule is the meaningful extra, though partial-failure and return behavior remain unstated.

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?

Two clauses in a single sentence, front-loading the capacity and upsert rule before the usage cue. Nothing is wasted or repeated from the annotations.

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 destructive, non-idempotent bulk write with no output schema, the definition covers capacity, the overwrite rule, and the primary use case, while annotations supply the safety profile. The remaining gap is what happens on partial failure or validation errors within a batch.

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?

At 67% schema coverage the schema carries most parameter detail, and the description only touches 'items' via the title-matching rule. 'brain' and 'returnIds' get no added meaning, and the nested item fields are deferred entirely to the schema's reference back to memory_upsert.

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?

States a concrete verb and resource ('saves ... memories'), plus scope ('up to 500 ... in one call') and the title-based upsert semantic. It is distinguishable from memory_upsert by the bulk framing, though it never names the sibling explicitly.

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?

'Use when the user wants to load many notes at once' gives a clear triggering condition. It stops short of an explicit when-not or routing statement (e.g. use memory_upsert for a single note), so the agent must infer the boundary itself.

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