Hi,
Stuck on MCP submission in the ChatGPT Apps platform. Looking to see if others hit the same wall.
After “Scan Tools” succeeds and all tools are detected, the form
throws:
Each MCP tool must set readOnlyHint, openWorldHint, and
destructiveHint to true or false.
Banner persists even after filling every justification field, blocking “Continue”.
Already verified
- All three annotations set as booleans in the server’s tool
definitions - MCP Inspector validates cleanly
- “Scan Tools” on OpenAI side detects every tool
- Justifications filled for every tool and hint
- Bumped version, cleared cache, refreshed MCP URL
Context
Found an existing GitHub issue from February with a March follow-up
reporting identical behavior. Still pending.
Questions
- Anyone submitted successfully in the last weeks?
- Platform-side bug or a specific annotation shape expected beyond
the MCP spec? - Known workaround?
Happy to share schemas in DM. Thanks
