本文へ移動
cccskills
無料GitHub で公開

subscription-recovery

Suede-owned recovery discipline for recurring charges billed outside Amazon: App Store, Google Play, PayPal, direct-bill streaming, gyms, news, and SaaS. Use when the user wants to find, audit, cancel, or dispute a subscription they may have forgotten, is being charged for twice, or no longer uses. Every cancellation and dispute requires the user to name the service first. Never enters payment credentials and never promises a refund. Requires Claude in Chrome for browser actions. NOT FOR: Amazon returns, restocking fees, or Amazon-billed Prime Video Channels, Audible, Kindle Unlimited, or Prime (use amazon-returns-recovery); merchant-side dunning and cancel-flow design (use suede-churn-prevention).

インストール方法を見る

含まれるファイル(4)

  • SKILL.md8.4 KB
  • agents/openai.yaml492 B
  • CARD.md4.8 KB
  • references/service-playbook.md1.6 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Subscription Recovery

IRON LAW: Never cancel, dispute, or contact a merchant about a service until the
user has named that specific service and the specific outcome they want for it.
A subscription appearing in a discovery list is not authorization to act on it.

Prerequisites

  • Claude in Chrome browser extension connected, for any service checked or acted on in-browser. If mcp__claude-in-chrome__* tools aren't loaded yet, fetch them via ToolSearch first.
  • Unlike amazon-returns-recovery, there is no single account to sweep, discovery depends on what the user can provide (a bank/card statement, app access, or just naming what they remember paying for) and what platform-level subscription hubs are available (App Store, Google Play, PayPal).
  • Unvalidated click-paths. Only the App Store, Google Play, and PayPal hub pages in Phase 1a are close to a fixed, checkable URL. Every individual service's own billing/cancellation page (Netflix, Spotify, a gym's member portal, etc.) has to be discovered live and should be recorded in references/service-playbook.md once confirmed, so the next run doesn't rediscover it from scratch.

Phase 1a: Platform subscription hubs (read-only, no side effects)

These three cover a large share of subscriptions in one page each, because the platform (not the individual service) is the merchant of record:

  1. Apple App Store (iOS/iPadOS/Mac subscriptions bought through Apple's in-app-purchase flow, many streaming apps route here instead of billing directly): https://apps.apple.com/account/subscriptions (requires Apple ID sign-in) or on-device: Settings → [Apple ID] → Subscriptions.
  2. Google Play: https://play.google.com/store/account/subscriptions, same idea for Android-purchased subscriptions.
  3. PayPal recurring payments: https://www.paypal.com/myaccount/autopay/: lists every merchant with standing authorization to charge the account, including ones that don't show up anywhere else (a common blind spot: an old free trial that converted, billed via PayPal, with no reminder email ever opened).

Each of these lists service name, price, billing cadence, and next charge date, and each has a direct cancel button on the same page: no negotiation needed for a straight cancellation found here.

Phase 1b: Bank/card statement scan (read-only, no side effects)

If the user can share a recent statement (PDF, CSV export, or even a screenshot of the transaction list), scan for recurring merchant names and amounts: the same charge appearing monthly/annually from the same merchant is the signal. This catches services that bill directly (Netflix, Hulu, Disney+, HBO Max, a gym, a SaaS tool) and aren't visible through the Phase 1a hubs. Ask for the statement rather than guessing; don't assume access to financial accounts.

Phase 1c: Ask directly

Ask the user what else they know they're paying for that Phase 1a/1b didn't surface: people usually remember 60-70% of their subscriptions when prompted but forget the rest until specifically asked. This is often faster than a statement scan for a first pass, and worth doing even after one.

Phase 1d: Amazon carve-out

If a subscription turns out to be Amazon-billed (Prime Video Channels, Audible, Kindle Unlimited, Prime itself), don't handle it here, hand off to amazon-returns-recovery's Phase 1b, which already documents those pages.

Phase 2: Confirm with the user

HALT. Discovery is over and nothing has been canceled, disputed, or contacted. Present the findings, then wait for the user. No service is acted on unless the user names it, silence, "sounds good," or a general go-ahead is not a naming.

Report every subscription found as a plain list: service, price, billing cadence, next charge date, and usage signal if known (last opened, last watched, last attended). Ask which ones to pursue and what outcome they want per service: cancel only, cancel and ask for the last charge back, or dispute a specific charge without canceling (e.g. billed twice in one month). Some subscriptions may turn out to still be wanted: don't assume every finding is a mistake, and say so if one looks intentional.

Phase 3: Execute (one service at a time, only after confirmation)

Straight cancellation, no negotiation needed:

  • If found via Phase 1a (App Store, Google Play, or PayPal), cancel directly on that same hub page, fastest path, no chat required.
  • Otherwise navigate to the service's own account/billing settings and look for a direct cancel option before resorting to chat or a phone call.

Refund/goodwill ask (forgot to cancel, charged after a cancellation attempt, billed twice, or genuinely unused for months):

  • Use the same ground rules as amazon-returns-recovery: state only true facts (service name, price, charge date, and the real reason), don't invent a prior contact attempt or cancellation date that didn't happen, ask plainly for the specific outcome wanted, and accept one polite counteroffer round at most before reporting back rather than escalating with anything untrue.
  • Apple and Google have their own self-service refund-request flows separate from the subscription hub itself, Apple: https://reportaproblem.apple.com; Google Play: order history → "Report a problem" (or support.google.com refund request). These are usually faster than chat for App Store/Play Store charges and worth trying first.
  • For direct-bill services, most have either a support chat or a cancellation retention flow (which sometimes offers a discount or partial refund unprompted when the user tries to cancel), take the retention offer only if the user actually wants to keep the service at the lower price; otherwise decline and proceed with cancellation.

Record the working path. Once a service's actual cancellation/dispute flow is confirmed live, add it to references/service-playbook.md with the exact URL and click-path, the same way amazon-returns-recovery documents its own chat flow: this is what turns "unvalidated" into "validated" over time.

Phase 4: Report

After each cancellation or dispute resolves, report per service: what happened (canceled, refunded, disputed and declined), the confirmation identifier or confirmation email, the effective end date, and the refund amount and method. An outcome with no confirmation identifier or email is reported as unconfirmed, never as done. For anything unresolved, name the exact next contact and date.

If several services were pursued in one session, summarize as a running total: keep one-time dollars recovered separate from ongoing monthly/annual savings from cancellations, since they're not the same kind of money.

Boundaries

  • Keep billing records, customer identifiers, personal reasons, and support transcripts in the user's private session. Public playbook updates may contain generic navigation only.

  • Never enter, store, or transcribe payment credentials, card numbers, or bank logins, and never connect a financial account.

  • Never promise a refund amount, a refund timeline, or that a dispute will succeed.

  • Never act on a service the user has not named, and never batch several services under one approval.

  • Never dispute or cancel a charge the user recognizes as intentional.

  • Never escalate with a fact the user did not supply: no invented prior contact, cancellation date, or usage claim.

Routing

  • Amazon-billed subscriptions, returns, and restocking fees -> use amazon-returns-recovery.
  • Designing the cancel flow, dunning, retention offers, or save offers for a product the user sells -> use suede-churn-prevention. That skill is the merchant side; this one is the subscriber side.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Suede-affiliated Amazon money-recovery audit for restocking fees, short or denied refunds, and Amazon-billed subscriptions. Use when the user mentions a restocking fee, a short or denied Amazon refund, or a forgotten Prime Video Channel, Audible, Kindle Unlimited, or Prime charge, or asks whether Amazon still owes or bills them. Read-only discovery; no chat, cancellation, or dispute without the account owner's per-item confirmation, and never a promise of recovery. Requires an authenticated Claude in Chrome session. NOT FOR: bank chargebacks, marketplace A-to-z seller claims, price-protection claims, or subscriptions billed outside Amazon (use subscription-recovery).

日本語の概要は準備中です。原文の説明を表示しています。

JasonColapietro/suede-creator-skills1272026年10月12日 更新

Suede Labs Android app factory: plan, build, test, and release a production-grade native Android app from a product idea through Google Play. Use for requests to create, ship, submit, monetize, or modernize an Android app. Covers Kotlin and Jetpack Compose architecture, API-level policy verification, accessibility, privacy and Data Safety, Play Billing, Play Integrity, account deletion, testing, performance, store assets, signing, staged rollout, and a release evidence gate. NOT FOR: iOS work (private Suede Labs companion, not in this pack: ios-app-factory), a review-only pass on an existing app (use suede-code-review), or live Play listing and keyword audits on a shipped app (use suede-aso).

日本語の概要は準備中です。原文の説明を表示しています。

JasonColapietro/suede-creator-skills1272026年10月12日 更新

Design, review, and visually QA web and app interfaces against explicit design laws so shipped screens look intentional instead of generic AI output.

日本語の概要は準備中です。原文の説明を表示しています。

JasonColapietro/suede-creator-skills1272026年10月12日 更新

Suede AI full-stack surface builder that runs design, copy, and visual QA as one pass: landing pages, brand surfaces, product UI, dashboards, campaigns, launch pages, and reference-to-target restyles (suedify). Use when a build needs layout and words together, when a redesign or launch surface has to ship end to end, when the request is 'make this site look like that one', or when design, copy, asset, and QA lanes have to move at once. NOT FOR: a design-token, dark-mode, or single-component decision (use suede-design); copy with no layout work (use johnny-suede-write, or suede-copy for one standalone conversion surface); conversion and funnel architecture (use suede-site-alchemy); multi-agent orchestration protocols, WIP protection, and rollback trees (use suede-agent-teams).

日本語の概要は準備中です。原文の説明を表示しています。

JasonColapietro/suede-creator-skills1272026年10月12日 更新

Suede Labs full writing stack: sharper copy for docs, pages, email, social, headlines, CTAs, product listings, and public explainers, with an SEO/AEO/AI EO pass, persona and framework selection, brand-voice alignment, and a scored ship gate. Use when a writing job spans more than one surface, needs a voice retune as well as a draft, needs discoverability metadata alongside the copy, when the document is one an agent reads such as a SKILL.md, CLAUDE.md, or AGENTS.md, or when the user asks for 'the full writing stack', a launch package, or a public explainer talk-track. NOT FOR: one standalone conversion surface in one pass (use suede-copy); stripping AI patterns from text you did not write (use suede-deslop); a researched multi-phase piece for a high-stakes public surface (use suede-ship-copy); a deep standalone SEO audit (use suede-seo-audit); copy that ships inside a design or layout build (use johnny-suede-design).

日本語の概要は準備中です。原文の説明を表示しています。

JasonColapietro/suede-creator-skills1272026年10月12日 更新

Suede Labs workflow for turning a website, PWA, dashboard, or marketplace into an iOS app. Use when the user has a live site or web app and asks to put it on the App Store, wrap it in an app, ship an iOS version, or convert a PWA to native, covers URL audit, shell-vs-native strategy, App Store 4.2 wrapper risk, native value requirements, screenshots, metadata, privacy answers, and the release gate. NOT FOR: building a native iOS app with no existing site (private Suede Labs companion, not in this pack: ios-swiftui-product); repairing or releasing an existing Capacitor shell (private Suede Labs companion, not in this pack: ios-capacitor-shell); Android conversions (use android-app-factory); live listing and keyword audits on a shipped app (use suede-aso).

日本語の概要は準備中です。原文の説明を表示しています。

JasonColapietro/suede-creator-skills1272026年10月12日 更新

JasonColapietro のスキルをすべて見る

このスキルの問題を報告する