Semantic Search

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

Annotations already declare readOnly/idempotent/openWorld, and the description adds genuinely useful context: the index is persisted, no implicit re-indexing happens, and the embedding cost is deliberate. It does not cover failure modes such as an empty or stale index, which is the main residual behavioral gap.

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?

Purpose and the key constraint are front-loaded, and every sentence carries information. The scoring-internals clause ('cosine similarity', 'top-chunk focus signal') is slightly more implementation detail than an agent needs to select or call the tool.

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?

With no output schema, the description still tells the agent what a result is (one per note, best chunk as snippet), the indexing prerequisite, and the natural next step. Nothing needed to call it correctly is missing.

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 100%, so the baseline is 3; the description earns extra by clarifying that ranking yields one result per note (so 'limit' counts notes, not chunks) and that the snippet comes from the best chunk. It still says nothing extra about 'folder' scope beyond the schema.

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 resource ('Search notes by meaning rather than keywords') and immediately contrasts with the keyword sibling search_notes, so the agent can pick correctly without opening a schema. The ranking model (best chunk per note) further pins down what comes back.

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?

Explicit prerequisites and follow-ups: run index_vault first, no auto-indexing because of embedding cost, and pair with get_note for full bodies. Together with the 'by meaning rather than keywords' framing, the when-to-use vs search_notes is unambiguous.

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