Ask
Read-only questions grounded in files, selections, diagnostics and workspace code.
PRODUCT / FORMERLY LOCALFORGE
Meet LOMVREN.
Build with your own AI.
A local-first autonomous engineering environment inside VS Code. Keep your model choice, compute and development workflow in your hands.
Active development · Product name and capabilities follow the public repository.
Plan
Inspect the architecture and create an implementation plan before changing code.
Agent
Prepare multi-file proposals, review diffs, validate approved edits and attempt repairs.
Your compute
Use local Ollama, configured OpenAI-compatible endpoints, or your own remote Ollama GPU over SSH.
Reviewed changes
Snapshot checks and atomic proposals help prevent stale changes from being applied.
Model management
Discover models, choose providers, and manage local downloads inside the extension.
Local first. Explicit destinations.
Local inference uses your configured runtime. Remote SSH mode sends prompts and included context to your selected host; API mode sends them to your configured endpoint. The extension does not operate a hosted inference service.
Terminal commands run with your existing account permissions. Reviewed file proposals and command approvals are part of the workflow; they do not provide an operating-system sandbox.
Read the product architecture ↗