forget

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

Discloses that the tool is destructive, specifies preserved memory types (caveat-layer, goal-layer, pinned), and mentions the dry_run capability via schema. However, the description does not detail the exact forgettingRisk formula or that deletion is irreversible, but the dry_run parameter is well-described in schema.

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 three sentences long, with the main action stated first. It is concise and free of unnecessary detail, efficiently communicating key information.

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?

Given the tool's complexity (two modes, preservation logic, destructive nature), the description covers essential aspects. It explains when to use each mode, what is preserved, and the destructive intent. No output schema is provided, but the description does not need to discuss return values as the tool's effect is primary.

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?

The description adds semantic context for memory_id (explicit deletion) but does not explicitly mention dry_run. However, it enriches the understanding by explaining the two operational modes and preservation rules, which go beyond the schema's parameter descriptions. Schema coverage is 50%, but the description compensates with broader context.

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 clearly states two distinct modes: explicit deletion by ID and auto-forgetting based on forgettingRisk. It also specifies preservation of caveat-layer, goal-layer, and pinned memories. This distinguishes it from siblings like update_memory and remember.

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?

Explicitly advises to prefer update_memory for corrections, indicating when not to use forget and providing an alternative. Implicitly guides when to use: for destructive deletion or auto-forgetting.

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