conductor-engine 0.8.0

Minimal orchestration runtime extracted from the Home AI Control Plane.

Conductor Engine

A minimal, installable orchestration runtime for task execution, capability loading, guardrails, storage abstractions, and future agent/policy layers.

What It Includes

• Generic task contracts and supervisor runtime
• Capability registry and plugin loading
• Built-in echo, filesystem, http, and optional memory capabilities
• Optional memU-backed memory provider abstraction
• Local JSON task store and in-memory queue
• cond CLI for local task execution and inspection
• Docs-first engine contracts under docs/conductor/

Quick Start

pip install -e .
cond capability list
cond run task.yaml
cond task list

Example task:

name: Echo smoke test
capability: echo
input:
  message: hello from conductor

Repository Layout

Conductor-Engine/
  engine/
  cli/
  docs/
  config/
  tests/

Built with Conductor Engine

home-ai-control-plane — A policy-governed, multi-agent AI control plane running on a Raspberry Pi 5. Manages personal digital workflows, home-lab services, and smart-home integrations with OPA-enforced approvals, budget limits, and a skill-based execution model. The system that motivated this engine.

→ How the engine maps to this use case

Automation

• .github/workflows/ci.yml validates Conventional Commit messages, runs ruff check ., and runs pytest tests/engine -q.
• .github/workflows/release.yml uses Python Semantic Release to calculate the next version, tag the release, build the package, and publish it to PyPI.
• PyPI publishing is configured for Trusted Publishing with the .github/workflows/release.yml workflow. On PyPI, register this repository and workflow as the trusted publisher. A GitHub environment is optional and is not required by the current workflow.
• If you want the current 0.1.0 in pyproject.toml to be the baseline release, create and push v0.1.0 before enabling the release workflow. Otherwise, semantic-release will calculate the next version from commit history.