aixai-labs

Official Python SDK for the AIxAI API.

AIxAI is an AI agency in Jaipur, India that builds and runs production AI systems: chatbots grounded in your own documents, workflow automation, data intelligence, and the pipelines underneath them.

The API is public and unauthenticated — there is no key to configure — and this package has no runtime dependencies.

pip install aixai-labs

The distribution is aixai-labs; the import is aixai.

Browse the service catalog

from aixai import AIxAI

client = AIxAI()

for service in client.iter_services():        # follows pagination for you
    print(service["slug"], "—", service["title"])

chatbots = client.get_service("custom-ai-chatbots")
print(chatbots["longDescription"])
print(chatbots["useCases"])

Find the right service for a need

for match in client.search("chatbot over our internal handbook"):
    print(match["score"], match["name"], match["url"])

Submit a project inquiry

Every non-dry-run call delivers a real email to the AIxAI team. Only send one with the end user's consent, using their own reply-to address.

# Rehearse first — validates the request without sending anything.
client.submit_inquiry(
    name="Ada Lovelace",
    email="ada@example.com",
    message="We want a RAG chatbot over our internal docs.",
    dry_run=True,
)

# Send for real.
result = client.submit_inquiry(
    name="Ada Lovelace",
    email="ada@example.com",
    message="We want a RAG chatbot over our internal docs.",
)
print(result["message"])

An Idempotency-Key is generated for you, so a retried request after a network failure replays the original response instead of delivering a second email. Pass idempotency_key=... to control it yourself.

There is also a dedicated sandbox that never delivers, whatever you pass:

client.sandbox_inquiry(name="Ada", email="ada@example.com", message="testing")

Errors

Every failure raises an exception carrying the API's structured error fields:

from aixai import AIxAI, RateLimitError, ValidationError, AIxAIError

try:
    client.submit_inquiry(name="Ada", email="not-an-email", message="hi")
except ValidationError as exc:
    print(exc.code, exc.message, exc.hint)     # validation_error ...
except RateLimitError as exc:
    print("wait", exc.retry_after, "seconds")
except AIxAIError as exc:
    print(exc.status, exc.code)

Rate limits are 10 requests/minute for inquiries and 60 for the catalog, per IP.

Command line

aixai services
aixai service custom-ai-chatbots
aixai search "automate invoice processing"
aixai contact --name "Ada Lovelace" --email ada@example.com --message "..." --dry-run

Other ways in

• MCP server — https://www.aixai.co.in/api/mcp (Streamable HTTP, no auth), plus a documentation server at /api/mcp-docs
• REST — OpenAPI specification
• Node CLI — aixai-cli
• Agent guidance — llms.txt

License

MIT. Contact: HQ@aixai.co.in