Update guardrails settings

Update the guardrails configuration for the authenticated user. Configure multiple guardrail modes including ban words (custom word list), prompt injection detection, and system prompt leakage detection. Each guardrail can be individually enabled/disabled with custom thresholds and rejection messages.

PUT /api/v1/llm/guardrails

Python (SDK)

Python

from mka1 import SDK


with SDK(
    bearer_auth="<YOUR_BEARER_TOKEN_HERE>",
) as sdk:

    res = sdk.guardrails.update_guardrails(guardrails=[
        {
            "mode": "ban_words",
            "config": {
                "words": [
                    "spam",
                    "inappropriate",
                ],
            },
        },
    ])

    # Handle response
    print(res)

200

{
  "object": "guardrails_settings",
  "guardrails": [
    {
      "mode": "ban_words",
      "enabled": true,
      "config": {
        "words": [
          "spam",
          "inappropriate"
        ]
      }
    }
  ],
  "updated_at": 1704067200
}

Authorizations

Authorization · string · header · required

Gateway auth: send Authorization: Bearer <mka1-api-key>. For multi-user server-side integrations, you can also send X-On-Behalf-Of: <external-user-id>.

Headers

X-On-Behalf-Of · string

Optional external end-user identifier forwarded by the API gateway.

Body

application/json

guardrails · object[] · required

List of guardrails to configure

Show child attributes

Response

200 - application/json

OK

object · any · required

guardrails · object[] · required

Show child attributes

updated_at · number · required

Unix timestamp of when settings were updated