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

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so safety is covered. The description adds useful behavioral context beyond that: the store contains historical and unconfirmed material, decisions are stored with the owner's exact wording, and unconfirmed notes exist but are excluded by default. It does not discuss result ordering or limits, though an output schema exists to carry the return shape.

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?

Content is front-loaded on the verb and resource, then supported by concrete example phrasings. It is a dense but single-purpose block with no filler sentences. Slight redundancy between the content list and the example queries keeps it off a 5.

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 zero-required-parameter search with an output schema present, the description covers what the store holds, why to search it, and that pending/unconfirmed entries exist. Combined with the schema's note that an omitted query lists the most important rows, an agent has enough to call it correctly. The only real gap is how this tool differs from overlapping siblings such as get_decision and get_learnings.

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 75% and the schema already documents query, account, and include_unconfirmed, so the baseline is 3. The description does add implicit meaning by enumerating memory content types (facts, preferences/decisions, attempts, outcomes, pending notes) that map onto the `kind` enum, giving the agent a reason to pick each value. It never names the parameters or explains how keyword matching interacts with filtering.

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?

Opens with a specific verb+resource ("Search what Taseli remembers") and scopes the content precisely: older facts, owner rules/decisions with exact words, attempts and their outcomes, unconfirmed notes. It is clear this is a search over persistent memory rather than a retrieval-by-id tool. It does not name the nearest siblings (get_learnings, list_decisions, get_decision, what_we_did), so the boundary is implied rather than stated.

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 concrete trigger phrasings — "what did I tell you about…", "did we try…", checking what the owner decided earlier — which tells an agent the situations that select this tool. It also implicitly excludes fresh/current business-card data by saying "beyond the business card." No explicit when-not or named-alternative routing is provided.

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