sas_prewarm

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

Annotations only declare readOnlyHint=false and destructiveHint=false; the description goes well beyond them by stating the call answers immediately (non-blocking background start), that no sign-in is ever prompted, the exact set of `state` values returned (starting / ready / idle / off / not_signed_in / unreachable / error), the step order performed (SingleStore library + startup code, then CAS), and two ways the feature can be disabled (connection settings or SINGLESTORE_MCP_SAS_PREWARM=0). This is exactly the kind of operational context structured fields cannot carry.

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 core action is front-loaded, but the whole description is a single long run-on sentence stitched together with parentheticals, backticked fragments, and an arrow-path UI reference. It reads as compressed internal documentation and requires a careful pass to parse, which costs more than it should for a two-parameter tool.

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 simple tool with two optional params and no output schema, the description covers behavior, non-blocking semantics, auth behavior, state enumeration, and opt-out mechanisms. What it leaves thin is the relationship to sibling state/status tools and the precise meaning of `connection` values, but nothing critical to invoking it correctly is missing.

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 description coverage is 0%, so the description must carry both parameters, and it largely does: `start` is explained as the two-mode switch (background start vs. check/keep-alive), and `connection` as defaulting to the active connection. It does not spell out how a connection is named/identified, and the default-true behavior of `start` is only implied by the phrasing, so it is not fully compensatory.

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 resource (the Workspace's SAS session for a SAS Viya connection) and a concrete effect: with `start` it launches Compute + CAS in the background so the first run doesn't wait, otherwise it polls a running session as a keep-alive. That is clearly more than a restatement of the name. It never names a sibling (e.g. `sas_run_status`, `sas_viya_state`, `sas_session_reset`) to sharpen the boundary, and the prose is dense enough that the verb/action distinction has to be inferred.

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?

Usage is implied rather than prescribed: prewarm before a first run, or call periodically as a keep-alive. There is no explicit statement of when NOT to use it, no mention of the overlapping siblings that report session state or reset sessions, and no guidance on how often to poll for keep-alive. The agent can infer intent but must decide between this and `sas_run_status`/`sas_viya_state` on its own.

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