Skip to content

quickstart.md

Latest commit

 

History

History
76 lines (52 loc) · 2.5 KB

quickstart.md

File metadata and controls

76 lines (52 loc) · 2.5 KB

5-Minute Quickstart

This is the fastest accurate path through the current MAPLE repo. It shows three things:

  1. a pure worldline example with no daemon dependency
  2. the daemon-backed runtime surface exposed by maple-cli
  3. the optional Ollama-backed playground path exposed by PALM

1. Build the tools

git clone https://github.com/mapleaiorg/maple.git
cd maple
cargo build -p maple-cli -p palm-daemon -p palm

2. Run a worldline example

cargo run --manifest-path examples/mwl-worldline-lifecycle/Cargo.toml

This is the quickest way to see the repo's core thesis in action: durable identity, lifecycle state, and recorded receipts instead of an ephemeral chat loop.

3. Start the daemon

In a new terminal:

cargo run -p maple-cli -- daemon start --foreground

4. Inspect the runtime from another terminal

cargo run -p maple-cli -- kernel status
cargo run -p maple-cli -- worldline create --profile agent --label quickstart-agent
cargo run -p maple-cli -- worldline list

Those commands exercise the current public runtime surface directly: kernel status, worldline creation, and daemon-backed control.

5. Optional: connect a local Ollama backend

If you want the local-model path too:

ollama serve
ollama pull llama3.2:3b

cargo run -p maple-cli -- doctor --model llama3.2:3b
cargo run -p palm -- playground set-backend \
  --kind local_llama \
  --model llama3.2:3b \
  --endpoint http://127.0.0.1:11434
cargo run -p palm -- playground infer "Summarize the current playground status"

This is the practical "Ollama-like" MAPLE workflow today: Ollama provides the local backend, PALM provides governed backend selection and inference entry points, and the model crates provide the underlying store, router, server, and benchmark types.

What you are seeing

  1. Worldlines turn actors into durable identities.
  2. PALM turns the runtime into a controllable daemon rather than a one-off process.
  3. maple-cli exposes current governance and provenance surfaces directly.
  4. Ollama is optional; governance and provenance remain visible without it.

If Ollama is not running

Skip step 5. The worldline, kernel, daemon, provenance, and governance flows are still valid and useful without a local model backend.

Next steps