sandbox_stop

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

With no annotations, the description carries the full burden and does a good job: it discloses recursive termination of child processes, the 8-slot capacity constraint, that output survives the stop and remains reachable via sandbox_check, and that the confirmation reports whether the command had already ended. The main undisclosed trait is whether stopping requires specific permissions or is reversible/restartable.

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, each with a job: what it does, when to use it, what it returns. The scoping to sandbox_start is front-loaded and there is 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?

An output schema exists, so the brief note about the confirmation return is a bonus rather than a requirement. The remaining gap is the undocumented parameters, which is material for a two-required-parameter mutation tool with zero schema coverage and no annotations.

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 explain both parameters and it does not name either one. An agent can guess that "the command" maps to job_id and that sandbox_id scopes it, but the relationship between the two identifiers and where job_id comes from (sandbox_start's response) is never stated.

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 ("Stop a background command") and immediately scopes it to commands created by sandbox_start, including recursive termination of child programs. This is enough to distinguish it from siblings like sandbox_check (which reads output) and sandbox_destroy (which presumably tears down the sandbox itself).

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 two concrete triggers: the command is no longer needed, or a background slot must be freed. It also routes the agent to sandbox_check for the output after stopping. What is missing is an explicit negative case, e.g. that this does not tear down the sandbox itself — that must be inferred from the sandbox_destroy sibling name.

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