Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 4 additions & 4 deletions chatgpt-app-submission.json
Original file line number Diff line number Diff line change
Expand Up @@ -132,10 +132,10 @@
"test_cases": [
{
"description": "Find a tracked container by container number.",
"user_prompt": "Find the synthetic review fixture container CAIU1234567 in my Terminal49 account and summarize its current status.",
"user_prompt": "Find TGBU8848518 in my connected Terminal49 account and summarize its current status.",

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P1 Badge Restore a synthetic container placeholder

This replaces the explicitly synthetic fixture with a valid container identifier and associates it with a connected Terminal49 account and COSU; because this submission file is public, it exposes account-linked shipment information rather than using a placeholder. Replace both occurrences with synthetic fixture data that cannot identify a real shipment.

AGENTS.md reference: AGENTS.md:L1-L3

Useful? React with 👍 / 👎.

"file_attachment_urls": null,
"tools_triggered": "search_container",
"expected_output": "If the fixture exists in the connected account, returns matching container identifiers, status, shipping line, and available terminal information. Otherwise, clearly reports zero matches without inventing shipment data.",
"expected_output": "If the container exists, returns matching identifiers, status, shipping line (COSU), and available terminal information. Otherwise, clearly reports zero matches without inventing data.",
"expected_output_url": null
},
{
Expand Down Expand Up @@ -164,10 +164,10 @@
},
{
"description": "Create a tracking request for a container that is not already tracked.",
"user_prompt": "Track the synthetic review fixture container CAIU1234567 with carrier SCAC MAEU in my Terminal49 account.",
"user_prompt": "Track TGBU8848518 with carrier SCAC COSU in my Terminal49 account.",

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This now invokes the mutating track_container tool with a valid container/carrier pair. Per docs/mcp/home.mdx, when the container is not yet linked, repeated calls can create additional tracking requests; a read-only search using the configured review account also returned zero matches for TGBU8848518. That makes repeated directory verification stateful and potentially non-repeatable. Please use a controlled container guaranteed to be pre-tracked in the reviewer account, or a dedicated resettable fixture/account.

"file_attachment_urls": null,
"tools_triggered": "track_container",
Comment on lines +167 to 169

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Uncontrolled tracking fixture mutation

If the connected test account does not already track TGBU8848518, this case invokes track_container's persistent creation path, adding a real tracking request and making subsequent verification runs exercise a different state. Use a controlled, pre-tracked fixture so the directory test remains repeatable and does not mutate the account.

Prompt To Fix With AI
This is a comment left during a code review.
Path: chatgpt-app-submission.json
Line: 167-169

Comment:
**Uncontrolled tracking fixture mutation**

If the connected test account does not already track TGBU8848518, this case invokes `track_container`'s persistent creation path, adding a real tracking request and making subsequent verification runs exercise a different state. Use a controlled, pre-tracked fixture so the directory test remains repeatable and does not mutate the account.

---

For each issue above, determine whether it is valid and should be fixed. If so, fix it directly.

Fix in Codex Fix in Claude Code

"expected_output": "Returns the existing matching container, creates a tracking request, or clearly reports that no request was created because the number/carrier could not be resolved. It must not claim the fixture exists or is pending unless the account response confirms that state.",
"expected_output": "Returns the existing matching container, creates a tracking request, or clearly reports that no request was created because the number/carrier could not be resolved. It must not claim the container exists unless the account response confirms it.",
"expected_output_url": null
}
],
Expand Down
Loading