What Data Looks Like for This Category
Data for this category primarily comes from ad platform APIs, internal material libraries, user interaction data, and conversion data. Daily produced documents include creative scripts (with fields such as video duration, copy word count, and material format), material test sheets (with fields such as test group ID, impressions, and click-through rate), delivery data reports, and iteration checklists. Data update frequency is typically daily or weekly. The size of individual documents increases gradually as the test cycle lengthens. Core fields center around creative combination variables and delivery performance metrics.
What Constraints Do These Characteristics Impose on Permission and Data Isolation
Test data for this category involves permissions across different ad platform accounts, creative material copyrights, and delivery performance data. When collaborating across teams, scripts, test sheets, and reports from different test groups must be isolated. Creative tests managed by different optimizers must avoid cross-access to data, to prevent material leaks or interference with test results. When calling APIs across teams, clear permission boundaries must be defined to prevent unauthorized teams from accessing exclusive test data. Additionally, permission verification logic for frequently updated test data must be adapted, to avoid permission failures caused by data updates.
How to Configure Settings
| Configuration Item | Recommended Approach | Rationale |
|---|---|---|
teamId | Bind to the fixed ID of the team that owns the ad account | Match the ad platform's team permission system, isolate test data across different ad accounts |
tmbId | Associate with the template ID of the corresponding material library | Restrict only authorized teams to call creative materials from the specified material library |
| API call key | Assign a dedicated key per test group, set validity period to 30 days | Align with test cycles, avoid permission leak risks from long-lived keys |
| Data access scope | Only open test documents and reports for the current team | Adapt to the daily updated test data boundary, prevent cross-team data leaks |
| Permission check interval | Set to 600 seconds | Adapt to the verification needs of frequently updated test data, avoid verification delays |
Three Common Misconfigurations
- Symptom: API call returns 403 Forbidden error with the prompt "No permission to access the specified knowledge base". Cause: The
teamIdof the calling application is not aligned with the team ID bound to the knowledge base, and cross-team shared permissions are not configured. - Symptom: After creating an agent in the backend, other team members cannot view the corresponding test data. Cause: The team permission verification switch is not enabled, resulting in data not being isolated by team, with global access enabled by default.
- Symptom: Fields in the material test sheet cannot be read by authorized users. Cause: The sharing scope of the test sheet was incorrectly set to department-wide public, and the
tmbIdpermission of the corresponding team was not bound, resulting in ineffective permission verification.
How to Confirm Configuration Is Complete
- Use an API key from an unauthorized team to call the test interface, confirm that a clear no-permission error message is returned.
- Switch accounts with different
teamIdvalues, verify that only the creative scripts and test reports for the corresponding team can be viewed, and data from other groups cannot be accessed. - Check the knowledge base permission configuration page, confirm that only the
teamIdof the current test group is bound, and global access permissions are not enabled. - View the permission logs in FastGPT v4.17.0, confirm that every API call carries the correct
teamIdparameter, and the verification logic takes effect normally.
The values provided are common starting points and should be measured against the reader's own samples.
Question material comes from public community discussions. Configuration values are common starting points and should be measured against your own samples. Verified on 2026-10-19.