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

Annotations give the safety profile (destructiveHint=true, readOnlyHint=false), and the description adds meaningful context beyond that: delete removes a version and its tags but sandboxes already started keep running; tag moves a tag rather than creating a new version. This is exactly the kind of side-effect disclosure the agent needs and couldn't infer from annotations alone.

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?

Purpose is front-loaded, then the four actions are laid out compactly with their parameter signatures. It is dense but every clause maps to an action or parameter, so little is wasted for a 4-action, 10-parameter tool.

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 and 10 parameters, the description does the needed work: it sketches what list and get return (newest-first images, state/version/tags/size/error/log lines) and covers the destructive semantics of tag and delete. It stops short of full return-shape or pagination-limit detail, but is complete enough to call correctly.

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 40%, so the description must compensate, and it does: it documents which parameters belong to each action (list: cursor/limit/state/name, get: id/logsAfter, tag: id/tag/remove), explains cursor round-tripping via nextCursor, and clarifies remove=true detaches a tag. Only idempotencyKey handling is left to the schema, which already documents it.

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 set of verbs (list, read, tag, delete) on a specific resource (the images sandboxes start from), and explicitly names the sibling responsible for creation: 'runtime_images_build makes them'. An agent can distinguish this from runtime_images_build and runtime_images_registries without opening a 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?

Each action is scoped with its own signature and behavior, and runtime_images_build is named as the alternative for building. It implies when each action applies (list to browse, get for a version's state/logs, tag to move a pointer, delete to remove) but does not state explicit when-not-to-use conditions or route to other siblings besides build.

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