household-capability-audit

A structured questionnaire that maps where your household is exposed: where you have genuine capability, where you have access masquerading as capability, and where you are simply dependent.

Part of the Solystopia open-source toolkit.

What it does

24 questions across 6 domains, scored against the four Solystopia capability questions:

Six domains assessed:

1. Compute & AI
2. Energy
3. Food & Water
4. Water
5. Connectivity
6. Fabrication & Repair

Output: a dependency profile with per-domain scores, a visual summary, and concrete next-action suggestions calibrated to where you are.

Install

No dependencies beyond Python 3.9+. No repo to clone required — pick one:

# As a CLI, straight from GitHub (no PyPI account needed)
pip install git+https://github.com/arthursilas-ai/household-capability-audit.git
household-capability-audit

# As an agent skill (Claude Code, Cursor, Copilot, and others)
npx skills add arthursilas-ai/household-capability-audit

# Or standalone — one file, no install
curl -O https://raw.githubusercontent.com/arthursilas-ai/household-capability-audit/main/audit.py
python audit.py

Usage

# Interactive questionnaire — all six domains
python audit.py

# Run with example answers to see what the output looks like
python audit.py --demo

# Assess specific domains only
python audit.py --domains compute energy

# Machine-readable output
python audit.py --json
python audit.py --demo --json

Example output

  HOUSEHOLD CAPABILITY AUDIT — RESULTS

  Domain                  Score   Level
  ────────────────────────────────────────────────────────────
  Compute & AI            ██░░  2.00  partially dependent
  Energy                  █░░░  1.00  fully dependent
  Food & Water            ██░░  2.00  partially dependent
  Water                   █░░░  1.00  fully dependent
  Connectivity            ██░░  1.75  partially dependent
  Fabrication & Repair    ███░  2.50  partially capable

  Overall score: 1.71 / 4.00

  Most exposed: Energy (1.00)
  Most capable: Fabrication & Repair (2.50)

  NEXT ACTIONS

  Compute & AI (partially dependent)
    · Run a local model once — llama.cpp + a GGUF file is the minimum viable step
    · Document what cloud services you depend on and what would break without them
    · Read the Sovereign AI Harness: github.com/arthursilas-ai/sovereign-ai-harness

The scoring scale

Context

This audit is the entry point for the Solystopia approach to household resilience. The four capability questions — knowledge, competence, substitutability, recovery — apply consistently across every domain. Each step you take in any domain moves one of those dials.

The audit was designed as the starting point referenced in the /tools page of solystopia.tech — the thing you do before you build.

Other tools in the Solystopia open-source ecosystem:

• hearthmind-economics — cost calculators for household compute
• agent-preflight — pre-deployment checks for AI agents
• piper-local-tts-demo — local text-to-speech
• sovereign-ai-harness — 10-stage local AI setup guide

License

MIT.