search

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

The description goes well beyond the annotations (readOnlyHint=true) by describing output trim (excerpts of ~120 chars), a tunable excerptLength, fuzzy matching and prefix search support, and the nuance that semantic/hybrid modes delegate to a local embedding model gated on SEEKSTONE_SEMANTIC=1. These details materially shape an agent's expectation of what the tool actually does when called.

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?

Three sentences, none wasted: purpose, the essential behavior (excerpts-not-notes), and mode/gating semantics. The most important constraint is front-loaded, and the rest earns its place by explaining mode behavior that the agent needs before calling.

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 complex search tool with no output schema, the description does enough heavy lifting: it tells the agent what the result looks like (exerpts), what modes exist, and how the mode-related feature gates work. The main gap is that it doesn't address result ranking, the absence of paging, or whether there is a structured query language — minor for the purpose, since the important return shape is specified.

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?

The schema already covers all six parameters at 100%, which sets a baseline of 3. The description adds meaning beyond the schema by explaining the query's inherent capabilities (fuzzy matching, prefix search), and by tying excerpt length tuning (~120 chars) to the excerptLength parameter's default. It doesn't restate schema field-by-field, so it earns the elevated score.

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?

The description clearly identifies the tool's action ('Full-text search across the vault') and adds a distinctive behavioral cue ('Returns ranked excerpts — not full notes'), which separates it from retrieval siblings like read_note without opening any schema. It stops short of explicitly naming a sibling to differentiate from, but the purpose is unambiguous.

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 description gives clear context for when this tool fits: it's a read-side discovery tool designed to minimize context usage, and it explains the mode trade-off (lexical vs semantic/hybrid, plus the SEEKSTONE_SEMANTIC=1 requirement). There is no instruction on when to use an alternative (e.g., query_notes or list_notes) and no explicit exclusions, so it earns a 4 rather than a 5.

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