The Autonomy Stack

The Autonomy Stack is an open, evolving governance operating system that offers a shared framework for designing legitimacy, coordination, and collective agency in a world where old institutions are no longer sufficient.

This repository exists for people who sense that governance matters deeply, yet rarely feels designed with care, imagination, or humility. It is a place to think together about how authority forms, how it holds, and how it might be re-made.

Why This Exists

Across societies, organizations, and communities, we are living through a widening gap between how systems of authority operate and how people actually experience them.

The Autonomy Stack begins from a simple conviction:
governance is something we build, maintain, and revise together.

Rather than offering answers in advance, this project creates space to explore better questions:

• What makes authority legitimate?
• How do people coordinate without forced agreement?
• How can systems remain coherent while allowing disagreement, dissent, and change?

How We Think About Governance

The Autonomy Stack treats governance as a form of design.

• Layered and modular
  Authority is shaped by many interacting components—rules, norms, tools, narratives, and practices—rather than a single structure.

• Iterative and adaptive
  Systems earn their credibility through use, revision, and responsiveness over time.

• Rooted in lived experience
  Legitimacy emerges when people recognize themselves in the systems they participate in.

• Capable of holding disagreement
  Durable governance does not require uniform belief, only enough shared structure to act together.

---

What You’ll Find Here

This repository is both a workshop and a record.

New to the project? Start here: stack/entry.md — a 2,000-word orientation for first-time readers.

• /stack/ The conceptual core. 00-overview.md is the current reference architecture (v0.3). STATE.md captures the current posture. LEXICON.md defines ~80 core terms. PROVOCATIONS.md holds the eight unresolved tensions the Stack must navigate. MODULE-CANDIDATES.md is the historical record of how the five modules were identified (all are now complete).

• /stack/modules/ Five completed modules applying the Stack to specific domains: Food Power, Care Infrastructure, Epistemic Coordination, Commons & Institutional Governance, Distributed Resistance.

• /stack/ practitioner documents entry.md — first-time reader orientation. design-toolkit.md — diagnostic questions by layer for practitioners. pattern-quick-reference.md — all 20 patterns in 1–2 sentences. agentic-governance-annex.md — cross-cutting analysis of governance under agentic conditions. tools/ — agentic governance toolkit for agent-facing diagnostics. cases/ — completed case studies.

• /stack/indexes/ Lightweight YAML indexes for agents navigating layers, patterns, and provocations. These route agents to source documents rather than replacing them.

• /patterns/ Twenty recurring governance dynamics distilled from practice. Patterns are discovered through the Metaviews: Future of Authority newsletter and stabilized here as reusable abstractions.

• /signals/ Short orientation artifacts (usually under 200 words). Timestamped. They surface shifts in attention, emerging tensions, and assumptions becoming visible — before they are ready to become patterns.

---

How to Engage

This work is meant to be read slowly and revisited often.

You are invited to:

• explore ideas without needing immediate conclusions
• contribute thoughts, critiques, and extensions
• treat uncertainty as a feature, not a flaw
• help shape something that none of us could design alone

Project Status

The Autonomy Stack has completed its first full roadmap execution across three maturation stages. The reference architecture (v0.3) has been grounded in eighteen months of applied research. Twenty patterns are documented. Five modules are complete. A lexicon (~80 terms), agentic governance annex, and full practitioner interface are in place.

The next roadmap direction is agentic governance tooling: making the framework usable by agents that build from open-source materials while preserving human accountability and contestability. The first draft toolkit is in stack/tools/, with an agent-readable index.yaml. See stack/ROADMAP-AGENTIC-GOVERNANCE-TOOLS.md and stack/STATE.md for the current orientation snapshot.

Validation

Run the documentation validator before committing changes to agent-facing indexes, tools, or source corpora:

python scripts/validate_docs.py

The validator checks local Markdown links, toolkit index consistency, required tool and protocol frontmatter, philosopher corpus paths, and source-corpus counts.

The work is alive, unfinished, and deliberately open. Its value lies not in arriving at certainty, but in cultivating clarity, shared language, and the capacity to act together under conditions of uncertainty.

If authority shapes our lives, then designing it is a responsibility worth taking seriously.