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

Annotations only carry destructiveHint=true, so the description must supply the operational picture — and it does, disclosing that the endpoint STARTS AUTOMATICALLY and bills per minute until stopped, and how to auto-stop it. It does not cover permissions, error modes, or provisioning latency, which keeps it short of a 5.

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?

Three tight clauses with the cost warning front-loaded and the resource-routing hint last. Every sentence earns its place with no filler.

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 an 8-parameter, nested-schema creation tool with no output schema, the description covers the highest-risk facts (billing, auto-start, auto-stop, hardware-id lookup). It omits nothing critical, though listing the initial-state STOPPED trick inline would have made it self-contained against the schema.

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 88%, so the schema already documents most parameters. The description adds real meaning for inactive_timeout (auto-stop semantics) and hardware id sourcing, but leaves autoscaling, availability_zone, and disable_speculative_decoding entirely to the schema. Baseline with minor added value.

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?

States a specific verb and resource ('Deploy a model on dedicated GPUs') and implicitly distinguishes creation from the sibling lifecycle tools together_start_endpoint and together_stop_endpoint. An agent can tell this creates a new endpoint rather than manipulating an existing one.

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 concrete routing guidance: 'use together_list_hardware for valid hardware ids' and 'set inactive_timeout to auto-stop it', plus the schema-level guidance to pass STOPPED to avoid billing. It does not explicitly contrast this with together_start_endpoint, but the create-vs-start distinction is clear from context.

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