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

Even though destructiveHint=true, idempotentHint=true, and readOnlyHint=false already exist in annotations, the description adds significant behavioral context: every memory entry is re-read at the start of each run, so stale entries cause false re-grounding; deleting a missing key returns deleted:false rather than an error; the 200-entry cap; scope limited to YOUR memory note; and tier restriction (sp500+). This goes well beyond the 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 front-loaded with the core action and use condition, followed by rationale, idempotency, capacity, scope, and tier. Every sentence adds information, though the tier line ('sp500+ (sample rejected)') is a bit cryptic and the whole text is denser than strictly necessary. Still well structured and not padded.

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 only one required parameter, a complete schema, an output schema, and annotations already covering destructive/idempotent, the description covers everything else: when to delete, why it matters, idempotent behavior on missing keys, the 200-entry cap, scope safety, and access tier. An agent lacks nothing needed to invoke this correctly.

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 100%, and the schema itself describes the key parameter well ('The memory key to forget. Discover keys with get_agent_memory (no key = list all).'). The tool description adds 'ONE durable' and 'your memory note' but does not materially extend the schema's parameter semantics. Baseline 3 is appropriate.

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 specific verb and resource: 'Forget ONE durable memory entry by key.' It also distinguishes from siblings by explicitly saying it never touches a thesis, claim, report, or financial fact, which separates it from delete_claim, delete_report, delete_thesis, etc. An agent can instantly tell this is for memory notes only.

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?

Gives explicit conditions for use: when a stored note is wrong, superseded, or was only scratch; when freeing a slot under the 200-entry cap. It also states what this tool does NOT do (never touches thesis/claim/report/financial fact), providing a clear when-not-to-use boundary against sibling tools.

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