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

Annotations already flag destructive/open-world/non-read-only, so the bar is lowered, but the description adds real behavior: start returns a URL plus Authorization header, secret settings map to Runtime secrets so the sandbox never holds the raw value, and servers honor sandbox network rules. This is useful beyond the annotations.

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?

The purpose leads, followed by a compact per-action breakdown and a closing note on secrets and networking. It is dense but every sentence contributes, and the most important information is front-loaded.

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 helpfully summarizes what start and get return and how secrets resolve. It is fairly complete for a multi-action tool, though it never addresses the undocumented port/replace parameters or the constraints of the nested servers configuration.

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 only 20%, so the description carries the compensation burden. It explains the `secrets` placeholder/egress-proxy semantics and implies `id` usage, but `port`, `replace`, and `options` are left completely undocumented in both schema and description, leaving significant gaps.

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 states a specific verb+resource: "Run well-known MCP servers in a sandbox," and then decomposes that into the four concrete operations (catalog/start/get/stop). It is clearly distinct from siblings like runtime_sandboxes_exec or runtime_sandboxes_create, though it does not name them to sharpen the contrast.

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?

Each action is given context: catalog lists servers and their hosts, start returns URLs and Authorization headers, get shows state and fresh URLs, stop ends them. That effectively guides action selection, but there is no explicit when-not or alternative-tool routing (e.g. vs runtime_sandboxes_create).

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