High signalMITOriginal

Drift Check

Verifies a project's reference data (prices, models, deps) against upstream sources — flags staleness.

Entry verified April 20, 2026

The short answer

Reference data rots. Model pricing changes, products get deprecated, APIs move. This skill checks your data files against their upstream source and flags drift before it shows up in user-facing pages.

When to use it

Weekly as a scheduled run. Before shipping a page that depends on pricing or model data.

Setup

  1. 1

    Save as ~/.claude/skills/drift-check.md.

  2. 2

    Configure source pairs: local-file + upstream-URL + fields-to-check.

  3. 3

    Can run as a GitHub Action on a cron schedule.

Example

You: /drift-check data/claude-models.json
Claude: 2 findings. Opus 4.7 price matches upstream ✓. Haiku 3 retiresOn=2026-04-19 is past (today 2026-04-20) — move to status=retired.

Source & attribution

Author
Bryan Collins (PWS pattern)
Licence
MIT
Type
Original

Original pattern published under MIT — attribution preserved by convention, not licence requirement.

Caveats

Upstream scraping breaks when sites restructure. Build fallback for HTML parse failures.

Related skills

Retire Model

Automates the lifecycle: current → legacy → deprecated → retired as dates pass. Keeps model data in sync with vendor announcements.

Pivot Sweep

Sweeps CTAs, URLs, or brand terms across a repo with tier-based policy — keep some, swap others, leave the rest.

Browse more skills

47 skills across 10 categories, all licence-verified.

Back to catalogueMore in Drift & regression monitoring
Read the guide