How do I distinguish ChatGPT research views from workspace tool access?
Montage’s compact research views and its workspace tools serve different purposes. When a ChatGPT mode exposes only part of the remote interface, diagnose that boundary rather than inventing a mapping between search results and account-scoped status execution. The comparison should name the missing workspace capability precisely. Compact research views may be useful without proving that a particular ChatGPT mode exposes signals_status or every account-scoped tool advertised by the server.
Help me distinguish ChatGPT research views from workspace tool access. Decide whether the task requires a compact research view or authenticated workspace status. Report the missing tool surface and use an approved compatible client if workspace status is required. No writes.Name the intended operation
Decide whether the task requires a compact research view or authenticated workspace status. A conversation that can find relevant material has not necessarily demonstrated access to every retained workspace operation. The task description should name which workspace fact is needed, since ordinary research results cannot prove account-specific workspace readiness.
Inspect mode availability
Review the tools the current ChatGPT mode actually offers for the selected app. Do not assume that a setup example for one mode promises the same tool inventory in every other conversation. A concrete inventory of offered tool names is better evidence than assuming an app that appears in settings is fully exposed everywhere.
Avoid synthetic tool equivalence
If signals_status is unavailable, do not ask search or fetch to impersonate it. Their output cannot prove the same account-scoped workspace facts simply because it comes through the same app. Search and fetch can be useful research interfaces without being substitutes for a missing status call; preserve that contract distinction.
Keep authorization independent
Even an available research view must follow its documented access rules. App selection and an OAuth grant should be evaluated separately from the mode’s ability to expose a particular tool. Do not add invented account parameters to a research request in an attempt to force it to act like a workspace operation.
Choose an honest next path
Report the missing tool surface and use an approved compatible client if workspace status is required. Do not claim a ChatGPT mode is broken merely because it supports a narrower research interface. The handoff should name the required missing capability so an approved alternate client can test that exact account-scoped read directly.
Keep the boundaries clear.
- Search and fetch can be useful research interfaces without being substitutes for a missing status call; preserve that contract distinction.
- The handoff should name the required missing capability so an approved alternate client can test that exact account-scoped read directly.
Start with a read-only check: ask your agent to report actual source coverage before collecting records or exporting an audience.
Primary references
Documentation reviewed 2026-11-15. Product capabilities, client availability and plans can change. Third-party marks identify their products, not an affiliation.