Red Team API Documentation
Red Team APIs
- GETRedteam Health
- POSTModel Health
- POSTModel Health of saved model
- POSTAdd Task
- POSTAdd Task with saved model
- POSTAdd Custom Task
- POSTAdd Custom Task with saved model
- PATCHCancel Task
- GETGet Task
- GETGet Task Status
- GETGet Task Results Summary
- GETGet Task Results Summary of a specific test type
- GETGet Task Results Details
- GETGet Task Results Details of a specific test type
- GETGet List of Tasks
Red Team APIs
Model Health
POST
/
redteam
/
model-health
curl --request POST \
--url https://api.enkryptai.com/redteam/model-health \
--header 'Content-Type: application/json' \
--header 'apikey: <api-key>' \
--data '{
"target_model_configuration": {
"model_name": "mistralai/Mistral-7B-Instruct-v0.1",
"testing_for": "Copilot",
"model_version": "v1",
"model_source": "https://together.ai",
"model_provider": "together",
"model_endpoint_url": "https://api.together.xyz/v1/chat/completions",
"model_api_key": "ABCDE12345678900",
"system_prompt": "",
"rate_per_min": 20,
"input_modalities": [
"text"
],
"output_modalities": [
"text"
],
"tools": [
{
"name": "web_search",
"description": "The tool web search is used to search the web for information related to finance."
}
]
}
}'
{
"status": "healthy",
"message": "Model health check completed successfully",
"data": {
"query": "What is the capital of India?",
"response": "The capital of India is New Delhi."
},
"error": "Error while checking model health"
}
Authorizations
Body
application/json
Example:
"https://api.together.xyz/v1/chat/completions"
Example:
"mistralai/Mistral-7B-Instruct-v0.1"
Available options:
foundationModels
, chatbotsAndCopilots
, agents
Example:
"Copilot"
Example:
"v1"
Example:
"https://together.ai"
Available options:
openai
, together
, huggingface
, groq
, azure_openai
, anthropic
, cohere
, bedrock
, gemini
, ai21
, fireworks
, alibaba
, portkey
, deepseek
, mistral
, llama
, openai_compatible
Example:
"together"
Example:
"ABCDE12345678900"
Example:
""
< 100 won't enable async. > 100 will enable async mode. > 200 we can run boosted async (all tests in parallel). Default 20.
Example:
20
Types of input that the model can process
Available options:
text
, image
, audio
Example:
["text"]
Types of output that the model can generate
Available options:
text
Example:
["text"]
Array of tools available to the model
Example:
[
{
"name": "web_search",
"description": "The tool web search is used to search the web for information related to finance."
}
]
Response
200 - application/json
Successful Response
Available options:
healthy
, unhealthy
Example:
"Model health check completed successfully"
Example:
"Error while checking model health"
curl --request POST \
--url https://api.enkryptai.com/redteam/model-health \
--header 'Content-Type: application/json' \
--header 'apikey: <api-key>' \
--data '{
"target_model_configuration": {
"model_name": "mistralai/Mistral-7B-Instruct-v0.1",
"testing_for": "Copilot",
"model_version": "v1",
"model_source": "https://together.ai",
"model_provider": "together",
"model_endpoint_url": "https://api.together.xyz/v1/chat/completions",
"model_api_key": "ABCDE12345678900",
"system_prompt": "",
"rate_per_min": 20,
"input_modalities": [
"text"
],
"output_modalities": [
"text"
],
"tools": [
{
"name": "web_search",
"description": "The tool web search is used to search the web for information related to finance."
}
]
}
}'
{
"status": "healthy",
"message": "Model health check completed successfully",
"data": {
"query": "What is the capital of India?",
"response": "The capital of India is New Delhi."
},
"error": "Error while checking model health"
}