Gather raw cited evidence of the user's own recurring language

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

Annotations only grant readOnlyHint=true; the description adds substantial context beyond that — restricting to role='user' words, spanning full history via brain, and explicitly disclaiming that it does not diagnose, summarize, or conclude. This is exactly the 'what it does not do' information annotations cannot carry.

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?

Front-loaded with the core scope and the non-conclusion disclaimer, then usage guidance. Slightly long and the downstream workflow caveat borders on instruction creep, but every sentence carries actionable content.

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?

An output schema exists, so return-value explanation is unnecessary. Given the read-only annotations, the description covers scope, the focus/omit behavior, and the intended follow-up workflow, leaving no material gap for correct invocation.

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 0%, so the description must compensate. It fully explains `focus` (a single angle vs. default battery of frustration/habit/drive/goal), but does not describe `limit_per_query`, leaving one of two parameters undocumented. Strong partial compensation.

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 — searches the user's own role='user' language across full transcript history to return cited evidence. It explicitly positions itself against siblings by noting it behaves 'same as goldfish_search' in returning excerpts, while previewing the downstream goldfish_remember step.

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 explicit when-to-use (behavioral/preference pattern evidence), when to pass `focus` (one specific angle) vs. omit it (default battery), and even prescribes downstream handling: write with goldfish_remember as tentative pattern-noticing and mention rarely. Nothing is left to inference.

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