Supersede memory

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

Annotations already declare readOnlyHint=false and destructiveHint=false, so the description's real job is the retention semantics, and it delivers: the old record remains queryable. It still omits whether the new memory receives a new id, what the call returns, and what happens when the referenced id doesn't exist.

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 sentences, zero filler, with the action first and the selection preference second. Every clause carries information the agent needs.

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 retention model is covered, but this is an 8-parameter mutation with 5 required fields, 13% schema coverage, and no output schema telling the agent what to expect back. The identity of the new record, optional-field behavior, and failure modes are left unaddressed.

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?

Only 13% of the 8 parameters have schema descriptions, so the description must carry meaning — and it does not. It never clarifies the critical split between the old record's id and the new memory's fields (title, body, kind, project), nor what happens to omitted optional fields like tags, source, or importance.

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States a specific verb and resource ("Replace an outdated memory with a new one") plus the distinguishing behavior: the old version survives and stays queryable. That contrast with plain deletion makes it immediately separable from memory_delete and memory_update without opening a 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?

Explicit routing rule: "Prefer this over delete whenever a fact changed" — a clear when-to-use with a named alternative. It never addresses memory_update, the closest sibling, leaving the update-vs-supersede choice to inference.

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