search

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

With no annotations, the description carries the burden and does disclose real behavior: output ordering (matching lifecycle facts first, then chunks), per-hit session_date with a required rendering template, a no-rewrite constraint on chunk text, and the conditional evidence second pass with its 12-note/6k cap and caching. It omits latency, permissions, and failure behavior, but the return shape is unusually well specified.

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?

The purpose is front-loaded, but the body drifts into implementation detail — RENOR_SEARCH_EVIDENCE, 'H10-v4 evidence second pass', '12-note/6k cap' — which is internal jargon an agent cannot act on. It also mixes formatting instructions into the same run-on block rather than separating routing from output handling.

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 2-parameter, annotation-free, no-output-schema tool, this covers purpose, sibling routing, result composition, and mandatory rendering conventions — enough to call and consume it correctly. The gaps are around comparisons to the remaining search siblings rather than missing mechanics.

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 both query and topK are already documented with type and range. The description implies the query style (natural-language who/what/where questions) but adds no format, syntax, or topK tuning guidance beyond the schema baseline.

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 role statement — the DEFAULT retrieval tool for questions about the user or the vault, scoped to identity/preference/project questions — and clarifies it returns lifecycle facts plus semantic chunks. It distinguishes itself from keyword search and from remember_fact, though it says nothing about how it differs from the evidence, entity, or facts siblings.

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 routing guidance: use this first for identity questions, do not start with keyword search, and send durable writes to remember_fact. It lacks exclusions or comparisons against evidence/entity/facts, which are the closest remaining alternatives, so it stops short of full when/when-not coverage.

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