Ask Tableau Agent

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

With annotations declaring destructiveHint=true, openWorldHint=true and non-idempotent, the description supplies the missing why: the agent can write semantic statements if asked, so questions must be phrased as questions. It also discloses latency ('can take a minute'), in-memory conversation lifetime tied to server restarts, and reliance on an undocumented web-app API that may change without notice — exactly the context annotations cannot convey.

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?

Purpose is front-loaded in the first clause, with mechanics and caveats following in short, ordered sentences. A little repetition around testing intent ('compare the numbers yourself' / 'fair way to test'), but no sentence is padding.

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, and the description still usefully explains the meaningful return fields (conversation_id, tools_used). Given the tool's side effects, latency and API fragility are all disclosed, the only real omission is the meaning of graph_id and what happens when it is null.

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 0%, so the description carries the load. It explains conversation_id thoroughly (follow-up vs. fresh, in-memory persistence) and implies question should be a single question, but graph_id is never mentioned at all, leaving one of three parameters undocumented anywhere.

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 and resource ('Ask Tableau's own agent ... a question about a graph and get its answer') and immediately frames the purpose as testing semantic context and Business Preferences against an answer key. That framing distinguishes it from siblings such as check_business_preferences or search_semantic_contexts, which inspect context rather than interrogate an agent.

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?

Gives clear operating guidance: ask one question at a time, compare numbers yourself, and omit conversation_id when you want a fresh conversation because that is 'the fair way to test'. It stops short of naming an explicit alternative tool or stating when this should not be used instead of the deterministic check tools.

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