Best for
- Use when the user wants to audit Amazon returns or Amazon-billed subscriptions, mentions a restocking fee, short or denied refund, forgotten Prime Video Channel, Audible, Kindle Unlimited, or Prime charge, or asks wheth…
JasonColapietro/suede-creator-skills/skills/amazon-returns-recovery/SKILL.md
Use when the user wants to audit Amazon returns or Amazon-billed subscriptions, mentions a restocking fee, short or denied refund, forgotten Prime Video Channel, Audible, Kindle Unlimited, or Prime charge, or asks whether Amazon still owes or bills them. Read order, refund, and subscription evidence first; report findings and require confirmation before chat, cancellation, or dispute. The documented cases recovered $448.31, but never promise recovery. Requires an authenticated Claude in Chrome s
Decision brief
Read order, refund, and subscription evidence first; report findings and require confirmation before chat, cancellation, or dispute. The documented cases recovered $448.
Compatibility matrix
| Platform | Status | Evidence | What to check |
|---|---|---|---|
| Codex | Not declared | No explicit evidence | Portability before use |
| Claude Code | Not declared | No explicit evidence | Portability before use |
| Cursor | Not declared | No explicit evidence | Portability before use |
| Gemini CLI | Not declared | No explicit evidence | Portability before use |
Installation
The source command is displayed only when detected. A safe inspection prompt is always available so your agent can explain every action before execution.
npx skills add https://github.com/JasonColapietro/suede-creator-skills --skill "skills/amazon-returns-recovery"Inspect the Agent Skill "amazon-returns-recovery" from https://github.com/JasonColapietro/suede-creator-skills/blob/f78965b9f99427cc4145950edbf9d388f38a3738/skills/amazon-returns-recovery/SKILL.md at commit f78965b9f99427cc4145950edbf9d388f38a3738. List every install step, command, network request, credential, file read/write, external action, and rollback step. Explain whether it fits my task. Do not install or execute anything until I approve.
Workflow
Goal: find every completed return where Amazon deducted a restocking fee, without touching anything.
Unvalidated click-path — the exact URLs below are the best-known entry points as of this writing, not yet confirmed live like the Phase 1a flow. If a URL 404s or redirects somewhere unexpected, navigate from the account menu instead (Amazon moves these pages periodically) and no…
Report the findings as a plain list: for fees, item / order / fee amount / who it shipped to / first-party or third-party; for subscriptions, name / price / billing cadence / next charge date / whether it looks used or forgotten. Ask which ones to pursue and what outcome they wa…
For fee/refund disputes, drive Amazon's live chat to request a waiver. The exact click-path, a critical popup-window workaround, and the escalation flow to a human associate are documented in references/dispute-chat-flow.md — read that file before starting this phase, since Amaz…
After each dispute or cancellation resolves (or if the associate declines), tell the user: amount recovered or subscription canceled, refund method, stated ETA or effective date, and associate name if given. If several items were pursued in one session, summarize as a running to…
Permission review
The documentation asks the agent to read local files, directories, or repositories.
[references/dispute-chat-flow.md](references/dispute-chat-flow.md) — read that fileEvidence record
| Signal | Value | Evidence type | Meaning |
|---|---|---|---|
| Quality score | 86/100 | Computed | Documentation, specificity, maintenance, and trust rules |
| Repository stars | 165 | Source | Repository attention, not individual Skill quality |
| Compatibility | 0 platforms | Source | Declared in the catalog source record |
| Usage guide | automated source guide | Editorial | Generated or reviewed according to the visible evidence level |
Pinned source
This started as a test of a general-purpose contract/customer-service negotiation tool — the goal was just to see if it could hold its own in a dispute. Pointed at a live Amazon account, it surfaced two restocking fees ($44.99 and $28.50) the account owner had no idea existed, on returns processed weeks earlier. Both got waived in the same sitting — the first refunded exactly the $44.99 fee, the second came back at $30.63 (the associate rounded up past the $28.50 fee as a goodwill gesture). Total recovered from fees alone: $75.62, on charges the account owner never would have gone looking for. The lesson: money like this doesn't announce itself — restocking fees, partial refunds, and price-protection differences sit quietly in order history unless something goes and checks. That's what this skill automates.
Treat this as a metal detector, not an autopilot: it surfaces what's buried and drafts the case, but every dispute goes through the account owner before anything gets sent.
The same account also had a third case worth knowing about, even though it's outside this skill's default restocking-fee scope: a $372.69 electric shaver the account owner had already been denied a refund on once — while still inside the return window. Re-disputed two days after the window closed, hands-off, the case ended in a full $372.69 refund with no return required — the associate just let the account owner keep it. Overturning an existing denial after the window shut is the ceiling of what a well-reasoned exception ask can get, not just fee waivers. See references/example-cases.md for the full writeup of all three real cases, including exactly what was said and what worked.
mcp__claude-in-chrome__* tools aren't
loaded yet, fetch them via ToolSearch first (see the extension's own MCP
instructions for the batch-load query).Goal: find every completed return where Amazon deducted a restocking fee, without touching anything.
your-orders/search matches loosely, so one keyword (e.g. "razor") surfaces
adjacent items (shavers, trimmers) too. Restocking fees concentrate on
higher-value electronics/appliances, so prioritize checking those over
consumables or clothing.
https://www.amazon.com/your-orders/search/ref=ppx_yo2ov_dt_b_search?opt=ab&search=<keyword>Your Returns.https://www.amazon.com/your-returns — but note it only shows roughly the last
3 months, so it's a supplement to the search sweep, not a replacement for it.https://www.amazon.com/your-orders/order-details?orderID=<orderID>
Find the Refund Total line — it has a small chevron that expands to an
itemized breakdown (Item(s) refund / Tax refund / Restocking fee / Refund Total).
Click it. Orders with no fee just show item + tax = refund total; orders with a
fee show the deduction explicitly.Stop here. Do not open the dispute chat yet.
Unvalidated click-path — the exact URLs below are the best-known entry points as of this writing, not yet confirmed live like the Phase 1a flow. If a URL 404s or redirects somewhere unexpected, navigate from the account menu instead (Amazon moves these pages periodically) and note the working path back into this file once confirmed.
Goal: find every recurring digital subscription billed through the Amazon account — Prime Video Channels, Audible, Kindle Unlimited, Prime itself — and flag ones that look forgotten, unused, or worth reconsidering. This is a different shape of "money Amazon is quietly taking" than restocking fees: it's ongoing, not a one-time deduction, so the fix is usually "cancel it" rather than "waive it," with a refund ask reserved for genuinely forgotten charges.
https://www.primevideo.com/settings/channels — lists every active channel
subscription, price, and next billing date. If that redirects, go to Prime
Video → Account & Settings (top right) → Channels.https://www.audible.com/account/membership-overview —
same Amazon login, separate billing page. Shows plan tier, price, next charge
date, and credit balance (unused credits are themselves worth flagging — they
don't expire immediately but do eventually).https://www.amazon.com/kindle-dbs/subscribe/kindle_unlimited
or via Account → Digital Services and Devices → Kindle Unlimited.https://www.amazon.com/manageprime — for cases
where the user isn't using Prime shipping/video/music benefits and it's worth
flagging (this one is more consequential to cancel than a $9 channel add-on, so
treat it as report-only unless the user specifically asks about it).Report the findings as a plain list: for fees, item / order # / fee amount / who it shipped to / first-party or third-party; for subscriptions, name / price / billing cadence / next charge date / whether it looks used or forgotten. Ask which ones to pursue and what outcome they want for each (waive a fee, dispute a charge, cancel a subscription, or cancel and ask for the last charge back). Some fees are legitimate (e.g. an opened-item policy the seller disclosed at return time) and some subscriptions may turn out to still be wanted — don't assume every finding is worth acting on, and say so if one looks earned or intentional rather than a mistake.
For fee/refund disputes, drive Amazon's live chat to request a waiver. The exact click-path, a critical popup-window workaround, and the escalation flow to a human associate are documented in references/dispute-chat-flow.md — read that file before starting this phase, since Amazon's chat UI has a specific gotcha (it opens in a popup window Claude in Chrome's tab tracking can't see) that will silently strand the flow if skipped. The same chat flow and associate-facing script apply whether the ask is "waive this restocking fee" or "cancel this channel and refund the last charge" — only the specifics of the ask change.
For subscription cancellations, note the two distinct asks are not equally strong:
Ground rules while chatting with the associate:
After each dispute or cancellation resolves (or if the associate declines), tell the user: amount recovered or subscription canceled, refund method, stated ETA or effective date, and associate name if given. If several items were pursued in one session, summarize as a running total across fees, refunds, and subscriptions canceled (report subscription savings as "$X/month going forward" separately from one-time dollars recovered — they're not the same kind of money).
Price-protection refunds: some categories/sellers refund the difference if a listing's price drops within the return window after purchase. This would mean comparing paid price vs. current listing price on recent (still-returnable) orders and flagging drops worth claiming. Worth building once the restocking-fee and subscription flows are both solid, but out of scope for now — don't attempt it without discussing it with the user first, since it's unvalidated.
Alternatives
coreyhaines31/marketingskills
When the user wants to plan, design, or implement an A/B test or experiment, or build a growth experimentation program. Also use when the user mentions "A/B test," "split test," "experiment," "test this change," "variant copy," "multivariate test," "hypothesis," "should I test this," "which version is better," "test two versions," "statistical significance," "how long should I run this test," "growth experiments," "experiment velocity," "experiment backlog," "ICE score," "experimentation program
coreyhaines31/marketingskills
When the user wants to reduce churn, build cancellation flows, set up save offers, recover failed payments, or implement retention strategies. Also use when the user mentions 'churn,' 'cancel flow,' 'offboarding,' 'save offer,' 'dunning,' 'failed payment recovery,' 'win-back,' 'retention,' 'exit survey,' 'pause subscription,' 'involuntary churn,' 'people keep canceling,' 'churn rate is too high,' 'how do I keep users,' or 'customers are leaving.' Use this whenever someone is losing subscribers o
event4u-app/agent-config
Use when the user says "review the design", "check the UI", or wants a comprehensive UI/UX review. Uses a 7-phase methodology covering interaction, responsiveness, accessibility, and more.
K-Dense-AI/scientific-agent-skills
Distributed computing for larger-than-RAM pandas/NumPy workflows. Use when you need to scale existing pandas/NumPy code beyond memory or across clusters. Best for parallel file processing, distributed ML, integration with existing pandas code. For out-of-core analytics on single machine use vaex; for in-memory speed use polars.