recall

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

With no annotations, the description carries the full responsibility for disclosing behavior. It reveals that the tool performs semantic search, returns facts with validity windows and receipts, excludes superseded facts by default, and requires quoting valid_from/invalid_at to avoid stale information. This is transparent about the core behavior and edge cases. It doesn't mention auth or side effects, but for a read-only search tool that is acceptable. The phrase 'receipts' hints at output but could be clearer.

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 description is four sentences long, each providing distinct value: it poses the purpose as a question, defines the tool's function, details the superseded-fact behavior, and gives a usage directive. It is front-loaded with the question that immediately clarifies intent. While it could be slightly tightened, it is not verbose and every sentence earns its place.

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 tool with no output schema and no annotations, the description covers the essential aspects: purpose, usage timing, and subtle behavior around validity windows and history. It does not describe the return format in detail (only mentions 'receipts'), nor does it explain the limit parameter, but these are minor gaps. The description is sufficient for an agent to invoke the tool correctly for its intended use case.

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 75%, so three parameters (query, as_of, include_history) have descriptions. The tool description adds meaning beyond that by explaining the concept of validity windows and the 'include_history' behavior, which clarifies the purpose of as_of and include_history. The limit parameter lacks a description in the schema and is not addressed in the tool description, which is a minor gap. Overall, the description enhances parameter understanding significantly.

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 starts with a clear question 'What has this brand's marketer already decided, corrected or rejected?' and then defines the tool as 'Semantic search over the brand's own memory, returning facts with their validity windows and receipts.' This is specific (semantic search over memory), identifies the resource (brand's memory), and distinguishes from sibling tools like 'get_brand_context' or 'get_account_memory' by focusing on past decisions and validity windows.

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 explicitly states 'Call it before proposing anything the marketer may have already ruled on,' which tells the agent when to use it. It also explains when to include history ('unless you ask for history') and instructs to 'quote valid_from/invalid_at rather than presenting an expired fact as current,' providing clear behavioral guidance. No alternatives are named, but the context makes it obvious this is the right tool for checking past rulings.

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