Scoped Memory Recall

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

The annotations include readOnlyHint: true and the description says 'Read-only and has no side effects,' but this is only a reinforced safeness; beyond that, it also adds what the annotation doesn't show: the action is 'semantic recall' over 'long-term storage', returning grounded facts, quotes, and source documents. It does not disclose rate limiting, token limits, or partial failures.

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?

Four crisp sentences. The first sentence packs the core capability; the second contains examples; the third shows the alternative; the fourth says read-only. No wasteful words.

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?

The data is a single high-level semantic retriever with a parameter count of 3 and a large number of sibling alternatives. The input schema and annotations are available; the output schema is presumably present (the signal indicates having an output schema). The description says facts/ citations, but there is no mention of 'no answer' style or edge cases. Existing output schema may resolve but is not shown.

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 has three properties overall with a 100% description coverage. The description words provide semantic context beyond schema: the user question is 'an input text, keyword, or concept'; top_k is 'the number of grounded facts to retrieve (default 10; min 1; max 50); space_id is 'optional partition area for restricting the memory. This does enough, and while the schema already reports defaults/min/max, its descriptions are not fully redundant with 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?

The description opens with a specific action ('performs associative semantic recall across long-term memory'), a precise resource ('to retrieve grounded facts, citations, and source documents'), and an explicit showcase of how it is distinct from its siblings: it calls out icx_find_facts for raw keyword matches and icx_quote_slice for character-exact verbatim quotes.

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?

States clearly when to use it ('when answering questions about already saved documents or facts'), exemplars included; the explicit alternatives are stated: for raw keyword matching across all entity names, use icx_find_facts instead; for exact verbatim quotes, use icx_quote_slice.

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