chatgpt_preview_changes

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

With annotations already declaring the mutation profile, the description still adds genuinely useful behavior: the 10-minute plan lifetime, the hard cap of 50 actions, and a non-obvious side-effect warning about shared-project visibility changes. Note the mild tension with readOnlyHint=false — the text says it does not modify ChatGPT, which is true of ChatGPT data but not of server-side plan state; an agent could misread this as a pure read. Not a hard contradiction, but worth flagging.

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 preview-only guarantee, then capabilities, then the risk caveat. Every clause carries information the agent needs and nothing is padded.

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?

There is no output schema, so the description should say what comes back — a plan identifier, an expiry, maybe a diff of pending changes — since that output is what chatgpt_execute_plan presumably consumes. It mentions the 10-minute window but not how the plan is referenced later, leaving the downstream workflow underspecified for what is inherently a two-step tool.

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 has to carry the burden, and it partially does by enumerating the five action kinds that map onto the oneOf branches. However it never explains the shape of the `actions` array, that each item is discriminated by a `type` constant, or the id patterns — all of which matter for constructing a valid call. The 'max 50 actions' note merely restates schema maxItems.

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 names a specific verb and artifact: it generates a cleanup plan valid for 10 minutes and explicitly does not modify ChatGPT. It then enumerates the supported operation types (rename, archive/restore, create/rename project, move conversation), so an agent knows the resource domain precisely. It loses the top mark only because it never names the companion tool that applies the plan (chatgpt_execute_plan), leaving that differentiation to inference from the word 'preview'.

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?

It gives one concrete usage prerequisite — verify the target project before moving conversations into a shared project, because visibility scope may change. That is real guidance. But it never states when to prefer this over chatgpt_execute_plan, or how the two-step preview-then-apply workflow is supposed to proceed, which is the central usage decision for this tool.

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