gcp-vertexai-generate-content-output
A model response can reproduce supplied secrets or generate sensitive tool-call arguments and output content.
Where it sits
| location | aiplatform.projects.locations.publishers.models.generateContent.candidates[].content.parts[] |
|---|
| location kind | output_value |
|---|
| data kinds | credential customer_data pii sensitive_data |
|---|
| emits edge | ContainsSecret |
|---|
| service | Vertex AI / Workbench (gcp:vertexai) |
|---|
Collection recipe
| access mode | data_plane |
|---|
| operation | aiplatform.projects.locations.publishers.models.generateContent |
|---|
| response path | candidates[].content.parts[] |
|---|
| encoding | json |
|---|
| params | {"name": "\u003cresource-name\u003e"} |
|---|
Required permissions
aiplatform.endpoints.predict
References