About GroundCortex
GroundCortex is an MCP server in the Files category: continuous LoRA fine-tuning as a local service. Watches source files for changes, retrains on change, and hot-swaps the adapter into a live OpenAI-compatible inference endpoint. Pipeline control via MCP.
Install
git clone https://github.com/huss-mo/GroundCortexThis server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.
Tools (4)
get_statusReturns the active adapter version, pending experience count, loaded adapters list, and last training run details.
Parameters not published.
list_adaptersList all successfully trained adapters with their version names and negative indices for easy switching.
Parameters not published.
switch_adapterActivate a previously trained adapter by version name, negative index (-1 = latest), or `"base"` to unload LoRA. Accepts `force=True` to load a `no-pass` adapter that failed the quality gate.
Parameters not published.
trigger_consolidationIngest all source files, train a new adapter if anything changed, and hot-swap it into the inference server. Returns the new version ID and training status.
Parameters not published.
Ask AI
Ask AI about GroundCortex
Powered by Claude · Grounded in docs
Security checks
6 checks have not been run yet.