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

Annotations already cover the safety profile (readOnlyHint=true, destructiveHint=false), and the description adds real behavioral context on top: chart facts are computed server-side, the shown signals are deterministic, the assistant must write the narrative fields itself, and those fields are labeled as AI interpretation. This meaningfully exceeds the annotation baseline, though it stops short of covering error or edge-case behavior.

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 sentences, each earning its place: purpose, contents displayed, computation/authorship contract, and post-call restraint rule. The embedded user phrase — 사주 궁합 봐주세요 — is borderline unnecessary but costs only a few words and arguably anchors the purpose context. No redundancy or filler otherwise.

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 rendering tool with nested inputs, an output schema, and high birth-field documentation inside the schema, the description is essentially sufficient: it tells the agent what the card shows, who computes what, how to compose the narrative, and how to behave after the call. The only notable gap is the lack of explicit routing between this and legend_saju_analyze_compatibility, plus how to proceed if no prior read_fortune call exists.

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?

With schema coverage at 33%, the description was expected to compensate; it does add meaning for the primary behaviors — birthA/birthA feed a server-side chart render, score is the assistant's editorial input, and narrative is assistant-authored AI interpretation. Still, several parameters (names and their formatting, gender/calendar nuances) are either self-evident or left to the schema, so the description doesn't fully close the coverage gap.

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 states a specific verb and resource: "Render a two-person compatibility card." It further distinguishes the tool by its exact contents — both day pillars and deterministic 합·충·형·파·해 signals — which separates it clearly from the single-person natal, fortune, and timeline card siblings in the tool list.

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 provides a workflow condition — the narrative must be grounded in a prior legend_saju_read_fortune call — and a post-call instruction to add only brief text the card doesn't show. However, it never explicitly states when to choose this tool over its analytical sibling legend_saju_analyze_compatibility or names any rejection condition, leaving some usage decisions to inference.

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