agentleFS
zombicozombico/mojulo/.claude/skills/figure-study/SKILL.md

figure-study

Skill4 starsChanged 33 days agoApache-2.0View on GitHub

Render the protoform human figure (male/female) as multi-view vexar studies — the whole body, or a focused piece like the bust — for visual analysis. Two modes: filled (lit mesh) and golden-ratio ring-wave wireframe (verify a region's construction before filling it). Also renders PROPORTION CASTS (figure-cast.js: heroic / brute / lithe / stout / child / chibi, or your own dials) as a contact sheet with a measurements table — use that after touching figure-cast.js, a cast preset, or any limb/segment length dial. Use when asked to "render/view/analyze the figure", "show the female form", "study the bust / <region>", "compare the casts / proportions / body types", to compare male|female, or after editing the figure builders. Invoke as `/figure-study <study-name>` or `/figure-study --casts [preset...]`.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Look at the protoform figure — as a whole or piece by piece — to analyze and tune its form.

    line 8
  2. Distinctive:

    Every preset (or the ones you name) × sex × front·¾·lateral, plus a dressed ¾ so the tailoring is judged on the same body, into one sheet — …

    line 41
  3. Very specific:

    Run this skill on <n>-<region>-study, Read it, tune, then fill.

    line 100

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

figure-study

Look at the protoform figure — as a whole or piece by piece — to analyze and tune its form. The figure is sculpted in the study spike and emitted as multi-view SVGs; this skill regenerates those and rasterizes the one you want so you can Read it.

Use it

node .claude/skills/figure-study/study.mjs <study> [<study>...] [--no-gen]

Runs the figure gen test (writes all study SVGs), rasterizes the named study(ies), and prints PNG paths. Then Read the printed PNGs — that's the analysis. Pass --no-gen to skip regen and just re-rasterize existing SVGs.

Examples:

  • … study.mjs 8a-stitched-vexar — the canonical male, frontal·¾·lateral.
  • … study.mjs 14-female-vexar — the female pole.
  • … study.mjs 13-dimorphism-pair — male | female side by side (the dimorphism axis).
  • … study.mjs 15-female-vexar-bust-study — a focused piece (wireframe wave-build).
  • … study.mjs 12-proto-sweep — the stockiness tuning sweep.

The cast study (proportions)

A separate entry point for the PROPORTION layer — cast (figure-cast.js) rather than the body builders. It renders through the production renderer, not the spike, so what you see is what a minted figure is.

node .claude/skills/figure-study/cast-study.mjs [<preset>...] [--cast '{…}'] [--proto '{…}']
                                                [--pose '{…}'] [--garment <key>] [--no-dressed]
                                                [--sex male|female] [--out dir]

Every preset (or the ones you name) × sex × front·¾·lateral, plus a dressed ¾ so the tailoring is judged on the same body, into one sheet — then Read the printed PNG. It also prints a table of height / leg / arm / ape-index / shoulderSpan per row, because each panel is auto-fit to its own figure: the sheet shows proportion honestly and stature not at all. Read the numbers for stature.

  • … cast-study.mjs — every preset, both poles (the full sheet).
  • … cast-study.mjs brute --proto '{"stockiness":1.35,"chestWidth":1.5,"bicep":1.6}' — the hulk read: proportion from the cast, mass from proto. Judge them together; neither is the figure alone.
  • … cast-study.mjs canonical --cast '{"forearm":1.4,"shoulderSpan":1.2}' — a custom cast beside the canonical figure, which is the comparison that tells you whether a dial did what you meant.
  • … cast-study.mjs chibi --proto '{"headScale":1.4}' — the mascot; chibi compresses the skeleton, headScale is what makes the head read big.

The studies (outputs)

All under lite-template/integration/0609/spike-output/figure-readable-envelope/, each a 3-view contact sheet (frontal · three-quarter · lateral) unless noted.

studywhat it shows
8-vexar / 8a-stitched-vexarcanonical male — basket vs stitched (pelvis folded, side/pec/scapula)
12-proto-sweepone figure at stockiness 0.82 / 1.0 / 1.22 (girth tuning)
13-dimorphism-pairmale | female (4 panels) — the sex axis from one generator
14-female-vexarthe female pole alone
15-female-vexar-bust-studypiece study: golden-ratio breast ring-waves, wireframe
48-human-buildssomatotype taxonomy — ecto / meso / endo × male / female (auto-fit → shape)
49-human-sizesstature lineup on a shared scale + common ground line (toddler→tall ♂); height reads true
50-child-head-ratioheadScale 1.0 / 1.4 / 1.8 ages one figure adult→child→toddler (heads-tall)

Two study modes

The figure is litFigureFaces(cam, positions, {stitched}) in the spike (figure-readable-envelope.spike.gen.test.js), posed neutral, rendered through the orbit camera. Front is az = Math.PI (the az=0 camera looks at the back — always label/aim from Math.PI).

  1. Filled (lit mesh) — litFacesFromRings(cam, ringStack, FLESH_HEX), vexar Lambert-shaded, back-face culled, depth-sorted. The whole body or a region. drawLitPanel auto-fits, so a panel fed only a region's rings zooms to it.
  2. Wave wireframe (verify before fill) — draw a region's surface as golden- ratio concentric ring-waves via projectLines + drawPanel (auto-fits → a zoomed piece study). Verify the wave layout closes the form and terminates at the right edge, then fill (mesh the same rings). This is how 15-bust-study works.

Recipe — add a focused piece study

To study a new region (e.g. a shoulder, glute, knee) the way the bust was done:

  1. Factor the region's surface into a point function in STAND space: regionPt(p, sgn, rr, a) → {x,y,z} — rr = footprint radius (0 = apex/centre → 1 = the terminator/outer edge), a = around-angle. (See breastPt.)
  2. Fill builder: regionRings stacks rings at uniform rr (each point toWorld(regionPt(...))), fed to litFacesFromRings.
  3. Verify builder: regionWaveLines emits concentric rings at golden-ratio rr (rr /= 1.618 each step from the terminator inward) as {pts, stroke} lines (STAND space — projectLines applies toWorld), plus faint radial spokes.
  4. New output: 3 orbit views (Math.PI, Math.PI+Math.PI/4, Math.PI+Math.PI/2), drawPanel([], projectLines(cam, lines), label, i*PANEL_W) — auto-zoomed. await fs.writeFile(path.join(OUTPUT_DIR, '<n>-<region>-study.svg'), …).
  5. Run this skill on <n>-<region>-study, Read it, tune, then fill.

Sex / tuning

DIM = DIMORPH.male | .female (set before a render) drives the dimorphic pole; GIRTH (stockiness), PROTO.height, and PROTO.headScale (skull size — the child↔adult lever) are the global knobs. See figure-proto-params.plan.md and figure-dimorphism.plan.md. The rig/metadata is figure-rig.js.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.