build_forecast

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

With no annotations, the description carries the full burden and does reasonably well: it discloses the caching short-circuit, the write-back side effect, the data dependencies (last seven days of news, dominant number, master-number flag, thirty-day activation window), and required infrastructure (Qdrant and an LLM endpoint). It omits failure modes and whether the write-back is idempotent or overwrites prior readings.

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?

Front-loaded with the purpose and caching rule in the first two sentences, then supporting detail. The long parenthetical listing data sources is dense but earns its place; no filler sentences.

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?

An output schema exists, so return values need not be described, and the description covers inputs, side effects, caching, and infrastructure prerequisites. What remains thin is the exact trigger conditions relative to sibling tools, which an agent might still have to guess.

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?

One required parameter 'day' with 0% schema description coverage; the schema only gives a date format. The description adds 'calendar day', which confirms the semantics but adds little beyond what the name and format already imply.

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 verb+resource ('Return the numerological reading for a calendar day') and immediately distinguishes itself from siblings like compute_numerology by stating it saves the result for later calls and assembles from stored news/activations rather than computing a bare number.

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 implies usage through the caching rule ('A day that was already read answers from Qdrant without running a model'), which tells the agent this is the read-through entry point. However, it never explicitly says when to prefer this over compute_numerology, get_history, or find_patterns, so routing is left to inference.

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