memory_forget

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

The description meaningfully refines the annotations: destructiveHint=true could imply irrecoverable deletion, but the description clarifies the operation is a tombstone append where prior history remains auditable — a genuinely non-obvious behavioral fact beyond what readOnlyHint=false/idempotentHint=true/destructiveHint=true convey. It stops short of explaining the effect on future recalls, but the core safety-and-mechanism profile is well covered.

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?

A single 10-word sentence with zero filler; the mechanism ('Append a tombstone') is front-loaded and the key behavioral property ('prior history remains auditable') follows immediately. Every phrase earns its place.

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?

The mechanism and auditability guarantee are covered, but the tool mutates state with no output schema and the agent is left without knowledge of return shape, the effect on subsequent memory_recall calls, or how the optional session_id/project_root parameters combine with the required scope/topic. For a 4-parameter mutation this is workable but has clear operational gaps.

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 0%, so the description bore the full burden of explaining scope (enum user/project/session), topic, session_id, and project_root — and it only touches 'topic' implicitly via 'for a topic'. The interplay of scope, session_id, and project_root is entirely undocumented, and the description fails to compensate for the schema's lack of param 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 states a specific mechanism — 'Append a tombstone for a topic' — with a clear verb (append) and resource (tombstone for a topic), and the auditability clause distinguishes this from a hard delete, which meaningfully clarifies what 'forget' does. However, 'tombstone' is jargon that isn't unpacked, and it doesn't explicitly differentiate from siblings like memory_remember or memory_recall.

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?

No guidance is given on when to use memory_forget versus any of its seven siblings, when forgetting is appropriate, or what prerequisites apply (e.g., whether the topic must exist, how scope selection affects behavior). An agent must infer usage entirely from the tool name and the one-line mechanism.

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