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

Annotations only declare the generic write/open-world/idempotent profile; the description adds real substance — that parameter sets are distributed across quoted nodes and do NOT parallelize a single solver, that auto-continuation can rent extra segments up to a budget, and that costs must be reviewed before confirming. This meaningfully extends the annotation profile. It does not explain why the tool is marked non-idempotent despite requiring an idempotency_key, which is a notable omission.

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 compact sentences, front-loaded with the paid/start action, then the distribution caveat, the auto-rental caveat, and a confirmation warning. No filler, though the clauses are dense enough that the parallelization warning reads slightly ambiguously.

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 13-parameter, deeply nested, paid, open-world mutation with no output schema, the description covers cost and scheduling behavior but leaves the majority of parameters and the return value entirely unexplained. An agent still needs to infer how to populate ssh, files, config, resume, and idempotency_key correctly.

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% across 13 parameters, so the description carries the burden and only partially delivers: it explains the quotes-as-nodes semantics, the auto budget/maxSegments rental behavior, and the three experiment variants. It says nothing about ssh, seed, files, level, config, resume, name, runs, or the required idempotency_key, leaving most 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?

The description names a specific verb ("Start") and the three concrete experiment types (preset, customer script, command), plus flags the operation as PAID. It is clearly distinguishable from listing/inspection siblings like emt_experiment_jobs or emt_monitor, though it never names an alternative outright.

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 stated: "Review all costs before confirming" hints that quotes should be obtained first (via emt_quote/emt_experiment_quote), but no sibling or prerequisite is named. There is no explicit when-not-to-use guidance, e.g. when to prefer emt_rent or emt_rent_cluster over a full experiment launch.

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