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

Annotations already declare the safety profile (readOnlyHint=false, destructiveHint=false, openWorldHint=false), so the description only needs to add context — and it does, disclosing that expired facts are retired automatically, which is lifecycle behavior the agent must reason about. The remaining gap is that idempotentHint=false means repeat writes may duplicate facts, and the description never warns about that or about what happens when content already exists.

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, front-loaded with the core action, with the conditional warning immediately after. No filler, and the emphasis on 'ALWAYS' is warranted because silent expiry loss is the main failure mode.

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 7-parameter write tool with full schema coverage and no output schema, the description covers the purpose and the one dangerous parameter well. The minor gap is that it never indicates what storing returns (an id, a confirmation) or how bank defaults behave, leaving the agent to infer the result of a non-idempotent write.

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 coverage is already 100%, so the baseline is 3; the description earns an extra point by explaining the intent behind the non-obvious 'snapshot' flag — a value that will change expires in 30 days — and tying it to 'expires' as an equivalent alternative. This adds meaning the schema's per-field text doesn't fully convey on its own.

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 object — 'Store a fact, rule or event in your team's memory' — which is far more precise than a tautology. It is clear about what the tool does, but with siblings like kb_write and kb_recall present, it never draws the distinction (why retain vs write?) for the agent.

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?

The second sentence gives a strong conditional rule for ephemeral data ('values that change ... ALWAYS set snapshot=true or an expiry'), which is real usage guidance. However, it says nothing about when this tool should be chosen over alternatives such as kb_write, and names no exclusions.

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