Remember a durable fact

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

Annotations already establish write/non-destructive/non-idempotent, but the description adds behavior not visible in structured fields: near-duplicates are refused by default, force overrides that refusal, and the fact propagates to brain_identity immediately. These are substantive operational traits (dedup semantics, side-effect timing) that an agent cannot infer from the schema or hints.

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?

Five short sentences, all load-bearing: action, content to include, override mechanism, and exclusion. The core action is front-loaded in the first clause and nothing is repeated from the schema or 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 5-parameter mutation tool with no output schema, the description covers purpose, required content guidance, duplicate handling, and scope limits. No output schema exists, so return-value explanation is not needed, and the annotations carry the safety profile.

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 only 40% (only force and category are documented), so the description must compensate. It does: 'Include why it matters and how to apply it' explains the why and how_to_apply fields, and the enumeration of fact kinds maps onto the category enum. It adds real meaning beyond the schema, though it never explains the 2000-character limit or the fact minLength constraint.

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+resource ('Append one durable fact about the owner to their memory file') and enumerates what counts as such a fact (preference, correction, project constraint, reference), which mirrors the category enum. The 'durable fact' framing plus the explicit exclusion of 'session trivia' effectively separates it from capture-style siblings without needing to name them.

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 a positive use case (durable facts), an explicit exclusion ('Do not store session trivia'), and a procedural alternative for the duplicate case ('pass force=true to store anyway'). It stops short of naming or routing to sibling tools for adjacent operations, so it is clear context rather than full when/when-not/alternative guidance.

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