Create sandbox

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

Adds substantial context beyond the annotations: billed cloud resources, returns an ID, 5-minute default lifetime, expiry deletes files, and no host files/env vars/API key are copied in (isolation guarantees). The 'do not retry blindly' warning aligns with idempotentHint=false. The expiry-deletes-files statement refers to the sandbox's own ephemeral contents rather than pre-existing user data, so it does not contradict destructiveHint=false.

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 tight sentences, front-loaded with the core action, then consequence/cost, then safety cautions. No filler; each sentence adds a distinct fact (cost, lifetime, isolation, approval/retry).

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?

With no output schema, the description correctly names the return value ('returns its ID') and covers cost, lifetime, isolation, and approval. It stops short of noting the lifetimeMs bounds or how the returned ID is consumed (e.g., by sandbox_exec/read_file), which would fully close the loop.

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 carries the param burden. It meaningfully interprets lifetimeMs ('default lifetime is 5 minutes') and implies the project constraint ('explicit project') and the confirm gate ('get the user's approval'), but it omits the 60s–60min bounds and never names the confirm flag directly. Partial compensation only.

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 ('Create an ephemeral interpreter sandbox') plus its scoping requirement ('in an explicit project'), which cleanly distinguishes it from sandbox_list, sandbox_show, sandbox_exec, and sandbox_delete. An agent can identify the operation without opening the schema.

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 operational gating: 'Get the user's approval before creating it; do not retry an uncertain response blindly.' That is strong pre-invocation guidance. However, it never names an alternative tool or states when *not* to create a sandbox (e.g., reuse an existing one from sandbox_list), so it stops short of explicit alternative routing.

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