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

Annotations only declare readOnlyHint=false, idempotentHint=true, destructiveHint=false. The description adds the behavior the annotations omit: the act is only valid inside a 30-day trash period after which the memory 'is erased for good', and the failure mode for an active memory. This is exactly the non-obvious state and error information an agent needs before calling a mutation.

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?

Three short sentences, front-loaded with the situation that triggers the tool, then the discovery step, then the constraint. Every sentence carries operational information; no restatement of the title.

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 4-param mutation with an output schema, the description covers the trigger, the id-discovery path, the validity window, and the rejection case. Return values need not be described because an output schema exists, so nothing material is missing.

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 already explains memory_id format and that reason is 'for the audit; ask the person if unsure', so the schema carries the load. The description contributes only an indirect hint that memory_id comes from list_memories 'forgotten' filtering; it says nothing extra about agent/run_id.

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 ('wants a forgotten memory back' → restore) and makes the scope explicit: only forgotten memories, not active ones. An agent can distinguish it from forget_memory, get_memory, and list_memories without opening any schema.

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 the triggering situation, the prerequisite discovery step ('Find its id with list_memories filter "forgotten"'), and an explicit when-not ('Restoring an active memory is rejected'), plus the 30-day window constraint. Nothing about routing or precondition checking is left to inference.

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