# Report a problem (/docs/support/report-a-problem)



## Include these details [#include-these-details]

* **Environment:** development or production.
* **Area:** extension, source, newsletter, reading list, EPUB library, delivery, account, billing, or
  admin.
* **Action:** what you selected immediately before the problem.
* **Expected result:** the outcome you thought would occur.
* **Actual result:** the complete visible message and status.
* **Reason code:** the stable code, including any detailed cause.
* **Time:** approximate date and time with time zone.
* **Client:** browser name and version, operating system, and extension version when relevant.
* **Public URL:** only when the source is already public and safe to share.
* **Reproduction:** the shortest sequence that produces the same result.

## Useful identifiers [#useful-identifiers]

An affected reading-list, device, book, or job identifier can help correlate privacy-safe operational
events. Prefer an opaque ID over copying the associated private content.

## Never include [#never-include]

* Passwords, passkey secrets, access tokens, authorization codes, or session cookies.
* Full extension callback URLs.
* Private subscription inbox or device delivery addresses.
* Payment-card information or unredacted Stripe screens.
* Private, paywalled, or copyrighted article bodies.
* Raw newsletter email unless an approved support process explicitly requests it.

## Screenshots [#screenshots]

Crop the image to the relevant interface, then redact names, addresses, bookmarks, browser profiles,
IDs that are not needed, and private titles. Follow the [screenshot checklist](/docs/support/screenshot-checklist).

## Parser failures [#parser-failures]

For a reproducible public page, include the canonical URL and reason code. This lets the team create a
test fixture and review a parser change. A failure does not automatically grant an AI agent production
access or cause an unreviewed pull request to be merged.