Skip to content
ZeroDirs

Generate listings

Markdown

.agents/skills/generate-listings/SKILL.md — 209 lines of instruction that ship inside the starter. This page is generated from it, so it says what the skill does without republishing it.

When an agent picks this up

Section titled “When an agent picks this up”

The skill description, verbatim — this is the text an agent matches your request against:

Fill a ZeroDirs directory with verified listings — the full pipeline from a candidates TSV to a promoted dataset. Verify every candidate exists with seed/check-links.mjs, write each description only from the fetched page, record provenance, lint the result with seed/check-listings.mjs, check the shape with seed/check-dataset.mjs –profile buyer, preview with ZERODIRS_DATASET, then promote with pnpm content:use. Use when asked to generate listings, write descriptions in bulk, seed a directory with content, or fill an empty directory site.

What it walks through

Section titled “What it walks through”
  1. Stage 1 — prove each candidate exists
  2. Stage 2 — write each listing from its own page
  3. Stage 3 — lint the prose
  4. Stage 4 — check the shape
  5. Stage 5 — preview without touching your content
  6. Stage 6 — promote
  7. Report back

Commands it runs

Section titled “Commands it runs”
Terminal window
node seed/check-links.mjs seed/.out/candidates.tsv --json seed/.out/links.json
node seed/check-listings.mjs seed/datasets/<slug>/listings
node seed/check-dataset.mjs seed/datasets/<slug> --profile buyer
pnpm content:use <slug> --yes
pnpm check && pnpm test && pnpm build
pnpm content:use

Files it touches

Section titled “Files it touches”

src/content/ · public/logos/ · src/content/schema.ts · site.config.ts · seed/datasets/ai-tools/ · seed/datasets/ai-tools/meta.json · seed/datasets/ai-tools/categories.json · seed/datasets/ai-tools/tags.json · seed/datasets/ai-tools/logos/ · seed/datasets/ai-tools/blog/ · src/loaders/files.ts · seed/datasets/

Using it

Section titled “Using it”

Skills load themselves — you describe the outcome, not the skill. “Add a Compliance category with an icon and seed it” is enough for an agent that can see .agents/skills/; Claude Code reads .claude/skills/, which pnpm skills:sync mirrors from .agents/ on every install. See Working with an agent.