Resume sandbox

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

Annotations cover the safety profile (readOnly=false, idempotent=true, destructive=true), and the description adds genuinely non-obvious behavior: it blocks for up to 2 minutes, the sandbox resumes with preserved memory/processes/files, and a failed resume preserves state rather than destroying it. It does not reconcile the destructiveHint=true flag with that 'state is kept' reassurance, leaving mild ambiguity about what is actually at risk.

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 short sentences, front-loaded with the action and its state-preservation guarantee, followed by the blocking behavior and the failure/retry note. No filler and nothing buried.

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 single-param tool with full schema coverage and annotations, this covers the operationally important facts: blocking duration, state preservation, and failure handling. It would be complete with a note on what a successful resume returns (status/connection info) since no output schema exists.

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?

Only one parameter (sandboxId) and schema coverage is 100%, with the schema itself explaining it is the exact ID from sandbox_create or sandbox_list. The description adds nothing about the parameter, so baseline 3 applies.

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 (start/resume) and resource (paused sandbox), and adds the crucial scope detail that it resumes 'where it left off: same memory, processes and files.' That framing cleanly distinguishes it from sandbox_pause, sandbox_fork, and workspace_resume without needing their schemas.

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?

'Start a paused sandbox again' implicitly scopes usage to paused sandboxes, and the closing 'If resuming fails, the saved state is kept; try again' gives explicit retry guidance. It stops short of naming an alternative (e.g., sandbox_create if the sandbox is gone) or stating what to do if the 2-minute wait times out.

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