jiazhizhong/higress
1
0
Fork 0
mirror of https://github.com/alibaba/higress.git synced 2026-05-27 22:27:29 +08:00
Code Issues Packages Projects Releases Wiki Activity

feat(ai-security-guard): add fallback JSON paths for response content extraction (#3738)

Browse Source 
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: rinfx <yucheng.lxr@alibaba-inc.com>
This commit is contained in:
JianweiWang
2026-04-28 14:58:59 +08:00
committed by GitHub
parent 1d33067372
commit 5173b4b2b8
6 changed files with 896 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files

30
plugins/wasm-go/extensions/ai-security-guard/README_EN.md
View File

@@ -29,6 +29,8 @@ Plugin Priority: `300`
| `requestContentJsonPath` | string | optional | `messages.@reverse.0.content` | Specify the jsonpath of the content to be detected in the request body |
| `responseContentJsonPath` | string | optional | `choices.0.message.content` | Specify the jsonpath of the content to be detected in the response body |
| `responseStreamContentJsonPath` | string | optional | `choices.0.delta.content` | Specify the jsonpath of the content to be detected in the streaming response body |
| `responseContentFallbackJsonPaths` | array | optional | [`choices.0.message.content`, `content.#(type=="text")#.text`] | Fallback paths tried in order when `responseContentJsonPath` extracts empty content; entries equal to the primary path are skipped automatically; set to `[]` to disable fallback explicitly |
| `responseStreamContentFallbackJsonPaths` | array | optional | [`choices.0.delta.content`, `delta.text`] | Streaming fallback paths tried in order when `responseStreamContentJsonPath` extracts empty content; entries equal to the primary path are skipped automatically; set to `[]` to disable fallback explicitly |
| `denyCode` | int | optional | 200 | Response status code when the specified content is illegal |
| `denyMessage` | string | optional | Drainage/non-streaming response in openai format, the answer content is the suggested answer from Alibaba Cloud content security | Response content when the specified content is illegal |
| `protocol` | string | optional | openai | protocol format, `openai` or `original` |
@@ -129,6 +131,34 @@ checkRequest: true
checkResponse: true
```
### Configure response fallback extraction paths
When primary extraction paths are empty, you can configure ordered fallback paths to support multiple response formats:
```yaml
serviceName: safecheck.dns
servicePort: 443
serviceHost: green-cip.cn-shanghai.aliyuncs.com
accessKey: "XXXXXXXXX"
secretKey: "XXXXXXXXXXXXXXX"
checkResponse: true
responseContentJsonPath: "choices.0.message.content"
responseStreamContentJsonPath: "choices.0.delta.content"
responseContentFallbackJsonPaths:
- "output.text"
- 'content.#(type=="text")#.text'
responseStreamContentFallbackJsonPaths:
- "payload.delta"
- "delta.text"
```
To enforce strict mode (no fallback), configure both fields as empty arrays:
```yaml
responseContentFallbackJsonPaths: []
responseStreamContentFallbackJsonPaths: []
```
## Observability
### Metric
ai-security-guard plugin provides following metrics: