opencad-agent

Natural-language CAD modelling on top of the opencad core.

The agent generates OpenCAD Python, validates it against an isolated analytic kernel, repairs it once if validation fails, then executes it — returning the generated code, the operations executed, and the updated feature tree.

Install:

pip install "opencad-agent[llm]"   # [llm] pulls LiteLLM for code generation

Use it against an existing runtime:

from opencad import reset_default_context
from opencad_agent import run_chat

ctx = reset_default_context()
response, operations = run_chat(ctx, "Create a mounting bracket with 4 standoffs")

Or drive the service directly:

from opencad_agent import OpenCadAgentService
from opencad_agent.models import ChatRequest

service = OpenCadAgentService()
result = service.chat(ChatRequest(message="Build a cog", tree_state=tree))

Configure the model with OPENCAD_LLM_MODEL and, when the provider needs it, OPENCAD_LLM_PROVIDER.

Kernel access

The agent never speaks HTTP. It takes a KernelClient (opencad.kernel.client) and the caller picks the transport — an in-process LocalKernelClient, or the HTTP client that opencad-backend supplies.

Tests

pytest

tests/agent/test_runtime_chat.py requires a configured live LLM; the rest of the suite runs offline with an injected completion function.