Ideonomy Skill — Development skill for Claude Code

Ideonomy — a skill for exploring an idea space

*Ideas are natural phenomena. They have properties. They live along dimensions. They can be sliced, negated, recombined, and re-instantiated like any other natural object.* — after Patrick Gunkel, via Grace Kind

A pair of skills that turn any idea into a populated neighborhood of related ideas, by applying combinations of *ideonomic operators* — negate, substitute, combine, abstract, re-instantiate, find-the-tree — drawn from a randomized tuple at every invocation.

Every call picks a different combination of operators, organons, and dimensional prompts from a catalog of millions of distinct method-tuples. The agent doesn't get to fall back on its default brainstorming moves. That's the whole point.

The two skills share one brain (operators × organons × dimension-prompts × picker × cooldown) and differ only in how they render the artifact:

Skill	Rendering	Best channel
ideonomy-plain	Plain Markdown lists, fenced-code-block tables, no Unicode box-drawing.	Anywhere — including SMS bridges, Telegram, plain Slack DMs, and any context where Unicode might mangle.
ideonomy-rich	Performative ASCII art: figlet banners, Unicode box-drawing, density gradients, visible ideonomy-machinery layers (tuple legend, dimensions surfaced, operator-named dividers, ideonomy trail).	Terminals, READMEs, blog posts with monospace code blocks, fixed-width-font emails, ttyrec sessions.

Default to `ideonomy-plain`. Use `ideonomy-rich` when you know the medium can hold the art.

---

Try the rite

$ ~/.claude/skills/ideonomy-plain/bin/pick    # or ideonomy-rich; same picker

=== IDEONOMY METHOD TUPLE (this invocation) ===

OPERATORS:
  - abstraction-lift
  - cross-domain-reinstantiation

ORGANONS:
  - periodic-grid

DIMENSION-PROMPTS:
  - longevity
  - autonomy
  - reversibility

==============================================

[…bodies of each picked method file follow, r