Decidealot MCP server

Decidealot

Your hardware. Local decision models. Run Laya and Von through TypeSafe-compatible HTTP or MCP.

At startup Decidealot downloads and verifies both local model bundles. It then loads only the model selected by a request, unloads it after the configured idle period, and returns typed choice, score, and noul answers with model probabilities. It exposes the TypeSafe HTTP API and MCP Streamable HTTP from the same local container.

Contents

• Quick start
• Use the API
• Use MCP
• Pick a model
• Configuration
• CUDA
• Model storage and unloading
• Agent integrations
• Docs

Quick start

You need Docker. This starts the CPU image on loopback, stores downloaded model files in one narrow host directory, and gives the container no capabilities or writable root filesystem.

model_directory="$HOME/.local/share/decidealot/models"
runtime_uid=$(id -u)
runtime_gid=$(id -g)
mkdir --parents "$model_directory"

docker run --detach --name decidealot --init --restart unless-stopped \
  --user "$runtime_uid:$runtime_gid" \
  --read-only --cap-drop ALL --security-opt no-new-privileges:true \
  --pids-limit 512 --memory 8g --cpus 4 \
  --tmpfs /tmp:rw,noexec,nosuid,size=128m \
  --tmpfs /var/run:rw,noexec,nosuid,size=8m \
  --log-driver json-file --log-opt max-size=10m --log-opt max-file=5 \
  --mount type=bind,source="$model_directory",target=/models \
  --publish 127.0.0.1:8080:8080 \
  psyb0t/decidealot:v0.4.0

Check the service, then ask Laya to make one typed decision:

curl --fail http://127.0.0.1:8080/health

curl --fail http://127.0.0.1:8080/v1/systemone \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "laya",
    "state": "A proposed action would permanently delete protected data.",
    "questions": {
      "handling": {
        "type": "choice",
        "instructions": "Choose the required handling for this action.",
        "criteria": {
          "allow": "The action is reversible and does not affect protected data.",
          "require_review": "The action is irreversible or affects protected data."
        }
      }
    }
  }'

The first service startup downloads both pinned model bundles and can take several minutes. Wait for /health before sending a decision. Later service starts reuse the same host directory. The response includes answers.handling.choice and a probability per choice key. Your caller chooses what to do with that decision, for example only allowing allow when its probability meets your own threshold.

Use the API

Send the state to judge and a bounded question. Decidealot returns typed results with probabilities.

Shortened. The full README is on GitHub.