acr-runtime 0.1.0

ACR — Adaptive Cognitive Runtime: a local-first, model-independent AI orchestration and cognitive runtime.

ACR — Adaptive Cognitive Runtime

Local-first, model-independent AI orchestration and cognitive runtime. Full specification: ACR_MASTER_SYSTEM_PROMPT.md. Current state and toolchain: docs/ARCHITECTURE.md — that document is the single source of truth for exactly what phase this repo is at; nothing here duplicates that status so it can't drift stale.

What's here

A Python CLI (acr) built around a real task engine, a hybrid-retrieval memory store, a skill registry with routing/validation/evolution, model and tool routing with safe-mode-aware permission checks, an agent planner/critic/topology system, an operational dashboard with a real-telemetry visualization, and an MCP server exposing ACR's memory, skill, web, and GitHub search tools to any MCP client. Runs entirely on SQLite with a zero-config mock model provider by default — no cloud account or API key required to try it. See docs/ARCHITECTURE.md for the full breakdown of what's implemented, how, and why.

Quick start

uv sync                 # install deps + local package into .venv
cp .env.example .env    # local dev data dir (repo-local ./data, gitignored)
uv run alembic upgrade head
uv run acr doctor

uv run acr run "say hello"          # create + execute a task end to end
uv run acr dashboard serve          # operational dashboard: http://127.0.0.1:8765
uv run acr mcp serve                # MCP server (stdio by default)

Using ACR from Claude Code

This repo ships a project-scoped .mcp.json, so opening it in Claude Code offers ACR's memory/skill/web/GitHub search tools and task execution as MCP tools directly — approve the prompt the first time it asks, then they're available every session.

Development

See CONTRIBUTING.md for the full workflow. Short version:

uv run pytest            # tests
uv run ruff check .      # lint
uv run ruff format .     # format
uv run pyright           # type check

License, security, support

MIT licensed. See SECURITY.md to report a vulnerability privately. For bugs, questions, or feature requests, use GitHub Issues.