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

Annotations already flag destructiveHint=true and openWorldHint=true, and the description adds meaningful context beyond that: the default call is a non-mutating preview, mutation requires confirm=true, and the preview surfaces changes made by other people. It stops short of saying whether a publish is reversible (e.g., rollback to a prior version) or what permissions are needed, which matters for an irreversible-looking live deploy.

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?

Two sentences, front-loaded with the action, then the safety protocol. Every clause is load-bearing – "including ones made by other people" and "after the user explicitly says yes in chat" both change agent behavior.

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 destructive, no-output-schema, 8-parameter deploy tool, the safety-critical flow is fully covered: preview semantics, the confirm gate, and the user-approval requirement. Gaps remain around what the preview response contains, what a successful publish returns, and container/workspace scoping, but none block a correct call.

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 63%, above the baseline threshold, and the description only adds meaning for confirm, describing its role in the preview-then-publish handshake. The other seven parameters (notes, detail, folder, language, container, workspace, version_name) get no mention in the description, so the agent must read the schema for those.

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?

"Publish the workspace draft to the live website" gives a specific verb (publish) and resource (workspace draft) plus the destination (live site). It is immediately distinguishable from read-only siblings like gtm_pending_changes or gtm_versions, which inspect rather than deploy.

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?

The description spells out an explicit two-step protocol: first call without confirm as a preview, show the user the full change list, then a second call with confirm=true only after explicit in-chat approval. That is a concrete when/when-not rule with no inference required.

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