agentleFS
zombicozombico/mojulo/.claude/skills/historic-culture/SKILL.md

historic-culture

Skill4 starsChanged 34 days agoApache-2.0View on GitHub

Add or deepen a historic culture (a real city or civilization at its period) in the historic-city generator — scaffold a new one at depth 0 on an existing culture's form, take one to its record and style card, its own assets or its own form and monument, add a region for a new part of the world, or land a culture branch on the release-candidate trunk. Use when asked to "add a civilization / city / culture", "start the Indus / Maya / Babylon …", "deepen <culture>", "give <culture> its own record / assets / form / monument", or "land / merge <culture>". Invoke as `/historic-culture <ask>`.

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:

    Read the guide first, every time: docs/historic/README.md.

    line 8
  2. Very specific:

    Run the gates (step 1). Mint it: create_sketch({ title, manifest: { kind: 'historic', culture: '<id>' } }).

    line 34
  3. Distinctive:

    Read its README's brief with them: what each relation offers, and the record entries that are parallels to verify (the first sources for …

    line 37

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

/historic-culture

Read the guide first, every time: docs/historic/README.md. It is the method (depths, the form and the solver, the asset loop, the card contract, the registries, the gates, landing on the trunk). This skill is the order of moves.

1. Where is it now?

cd control
npx vitest run lib/graph/historic/cultures.test.js lib/graph/historic/entries.test.js

For an existing culture, read its entry's DEPTH and BASIS lines (lib/graph/historic/entries.js entryCard('<id>'), or get_view_vocab({ id: '<id>' }) through the MCP). The next step is the next depth, never a skip.

2. A new culture: depth 0

  1. Settle with the operator: the place, the period and a read-at year (or none: a general depiction of the span), the region (a new row in lib/graph/historic/regions.js if it is a new part of the world), which existing culture's form has the town's SHAPE (the guide's table: its site, water and armature), and its LINEAGE: which registered cultures it continues, inherits from, took forms from by contact, or varies (get_view_vocab({ id: 'historic-lineage' }) or lib/graph/historic/lineage.js). Propose them; the operator corrects.
  2. Dry run, then write:
    node scripts/new-culture.mjs <id> --like <culture> --label "…" --years <from>,<to> --period "…" \
      --place "…" --region <region> --aliases "…" --draws <culture>:<kind>,… --dry
  3. Run the gates (step 1). Mint it: create_sketch({ title, manifest: { kind: 'historic', culture: '<id>' } }).
  4. Tell the operator plainly: at depth 0 it is the parent's town in the parent's colours; the palette is next. Its field guide (the mint's guideUrl) shows every choice the solver made and the knob for each. Read its README's brief with them: what each relation offers, and the record entries that are parallels to verify (the first sources for its own record at depth 1).

3. Deepen

  • Palette and skins from reference images (indexed in docs/historic/<id>/).
  • Depth 1: record/<id>.js (cited; confidence read / secondary / unverified; disputes; anachronisms held at their dates) and style/<id>.js (a machine check per principle in style.test.js); record on the card.
  • Depth 2: the asset loop, one asset at a time (the solver's slots → assetCall → sheet → assetBlueprint → kit → place → review). Never one-shot the town.
  • Depth 3: its own form, a row in form/towns.js (site, water, armature, enclosure, streets, density, placement, hinterland, activities), each principle citing its record's rows (cites), and its monument in monuments/ (one building, at the record's size); drop form from the card. Read the field guide and the anachronism check (ledger/) against its year: a principle or an activity whose rows are not yet in reach is a finding.

Each step: write its plan and its CHANGELOG ### section first (CLAUDE.md), keep every other culture's town as it was (its form and its solved plan), and stop at the eyes gate for the operator. Never claim the eyes gate passed.

4. Land it on the trunk

Merge the release-candidate trunk into the culture's branch first; registries and the changelog keep both sides; run the gates and the other cultures' page hashes; then the trunk merges the branch (no side branch left).

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.