Index of every page on this site: llms.txt
All skills
wix avatar

/rp-quick-runtime

@e2bc48e official
by Wix.comwix/skills40 stars
42
Run it yourself

Shared deterministic target-limit recovery and final-report runtime every rp-quick-<platform> quick-mode adapter uses before a Wix Stores write.

  • 3 files
  • 13.2 KB
  • Updated last week
  • GitHub
Use this Skill: https://skilld.dev/gh/wix/skills/rp-quick-runtime

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈41 tokens for metadata: the name and description. ≈945 when used: this file.

Description uses 2.0% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈163 characters, or 2.0%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

rp-quick-runtime

Not invoked directly — this is a shared library (lib/limit-recovery.js), not a standalone adapter. Every rp-quick-<platform> quick-mode adapter (rp-quick-shopify, rp-quick-woocommerce) requires it from its own scripts/quick-mode.js and must use it from its first version. Platform adapters own source extraction and field semantics; this runtime owns target-limit enforcement, recovery-ledger aggregation, and the human-readable final report — so a Wix Stores validation limit never silently turns an otherwise-valid record into an opaque skip.

const recovery = require('../rp-quick-runtime/lib/limit-recovery.js');

const normalized = recovery.normalizeProduct({ sourceId, sku, tagIds });
// normalized.sku / normalized.tagIds are the capped/trimmed values to write;
// normalized.recoveries lists what was capped or trimmed, if anything.
await writeProduct({ ...product, sku: normalized.sku, tagIds: normalized.tagIds });
// Record written_with_loss only after the target write succeeds, and never in a dry run.
if (normalized.recoveries.length) {
  if (!dryRun) await recovery.appendRecoveries(projectDir, adapterName, sourceId, normalized.recoveries);
  recovery.recordRecoveredRecord(summary, normalized.recoveries);
}
// ...after the run:
await recovery.writeFinalReport(projectDir, summary, { dryRun });

normalizeProduct applies the two verified Stores limits deterministically: caps tagIds at STORES_PRODUCT_TAGS_MAX (first-seen source order, extras omitted) and trims sku to STORES_SKU_MAX Unicode-safe characters with a stable source-ID-derived suffix so a trimmed SKU stays unique. Every cap/trim is reported back in recoveries[], never applied silently.

appendRecoveries ledgers each recovery to execution/quick-limit-recoveries.ndjson under the adapter's own name, tagged outcome: written_with_loss — the durable record of what was preserved versus lost, per source record. Call it only after the target write succeeds.

recordRecoveredRecord increments summary.recovered once for the written source record and increments summary.recoveryCounts once for each recovery event on that record.

writeFinalReport renders execution/quick-mode-final-report.md: a count of records imported with deterministic recovery plus a per-recovery-code breakdown, so the operator gets a compact, actionable account instead of a bare skip count. It adds a "Collections not migrated" section from summary.excludedCollections, a "Products to check" section from summary.writtenUnverified.product, and an inventory pointer when summary.inventory is set. With { dryRun: true } it writes execution/dry-run-final-report.md instead and sets summary.finalReport to that path, so a dry run can never overwrite or stand in for the live report. A dry run must write no live artifact at all (report, ledger, audit log, crosswalk or marker); adapters skip those writes when dryRun is true. writeFileAtomic (temp file, then rename) is the shared way to publish a report that a crash must not leave half-written.

lib/import-progress.js holds three more shared pieces. Each adapter's import first calls assertQuickImportAdapter, which fails unless the plan and extract manifest both carry exactly the loaded contract's { id, version, fingerprint }. It then calls writeQuickImportStartedMarker(projectDir, plan.adapter) before its first destination write. It is skipped on a dry run and never rewritten on resume; concurrent first attempts publish one marker (temp file, then a no-replace hard link). It writes state/quick-import-started.json, which the router reads as evidence of an interrupted import. validateQuickExecutionManifest is the router's check of the manifest that plan writes.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No third-party reports yet.

Signed by skilld at e2bc48e. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated last week

README badge

README badge for wix/skills/rp-quick-runtime