What the documents look like
Title and detail compliance deals with advertising truthfulness and qualification display. It does not deal with parcel delivery. The main data sources are the platform rules centre, the advertising law compliance checklist, the brand asset library, and third-party inspection reports. Everyday documents include the product title table, detail page assets, the prohibited-terms list, and inspection reports. The three most time-consuming steps are checking absolute terms, comparing declared parameters against the physical item, and replacing non-compliant assets.
These characteristics shape what conversation logs do at this step. A title or detail review produces a conversation identity, run details, and a record of which fields were retained. Operating reports and exception reviews compare those records against the underlying business documents. The stated update frequency is weekly and on each new listing; this page's function does not depend on that frequency, so it is not expanded here.
Constraints at the conversation logs step
Conversation identity must be checked against chatId together with the business record. Sessions from different entry points are not counted together. When run details are needed, set detail according to the official contract and inspect responseData. The log and the fields actually returned by the interface are the only basis for checking; business data that was not returned cannot be assumed to have been retained in full. Identity fields in test mode and in production calls may differ, so inputs, history, and returned content are checked separately. Reconciliation, retention, and compliance requirements are handled by a complete business plan. The existence of a log does not mean a compliance audit is complete.
Hypothetical paper exercise. The material text to be checked reads: "Product title table row 12; subject is the store operations role; object is a certain insulated bottle title; status is pending compliance re-review; date scope is the three titles listed in the week of October 2024; authorisation boundary is limited to compliance checking of this store's titles and detail assets." The request to be judged reads: "Please export the review conversation record for this insulated bottle title for this week's exception review." The authorisation boundary in the material does not state whether export is included, so it remains to be confirmed.
Configuration and implementation checks
The table below is built around the exercise above. It lists only settings or implementation checks that the official documents describe and that relate to this function and this primary task.
| Configuration or check | Hypothetical starting value for this exercise | Reason and comparison object |
|---|---|---|
| appId | The exercise application ID | The official text states appId identifies the application and works with chatId to locate a session |
| chatId | The exercise session ID | The official text states a non-empty chatId retrieves the session from the database, must be unique, and must be under 250 characters |
| detail | true | The official text states detail returns intermediate values and the full result, which is needed to obtain responseData |
| stream | false | The official text states that in non-stream mode the result is kept in responseData, which allows field-by-field checking |
| source | api | The official text states source=api retrieves only sessions created through the API, separating them from page sessions |
| loadCustomFeedbacks | true | The official text states this optional parameter reads custom feedback, used to check feedback fields |
Three common errors
- Counting sessions from different entry points together. Input is the exercise chatId and a record from a page session. Compare whether the two chatIds match. If they do not match and the records are still merged into one count, the conversation identity check fails.
- Claiming run details were obtained when detail was not set. Input is the exercise request with detail absent or false. Compare whether responseData appears in the return. If it does not appear and the report still lists module run details, that conclusion is not supported.
- Treating the existence of a log as a completed compliance audit. Input is the exercise conversation record and the compliance audit requirements. Compare whether the record covers reconciliation, retention, and the authorisation boundary. If only the conversation record exists and the authorisation boundary remains to be confirmed, the audit cannot be judged complete.
How to verify the setup
- Input the exercise chatId and a session list request with source=api. Compare whether the chatId in the returned list matches the exercise chatId. If it matches and total agrees with the expected session count, the conversation identity check is satisfied.
- Input the exercise request with detail=true and stream=false. Compare whether the returned responseData contains fields such as module name, running time, and tokens. If those fields are present and correspond to the requested period, the run detail retrieval is satisfied.
- Input the exercise request to be judged. Compare whether the authorisation boundary in the material states export. If the material does not state it, record this as an evidence gap and do not judge that export has been authorised.