openclaudia/openclaudia-skills/stripe-dispute
Claude Code and agent skill by openclaudia/openclaudia-skills
- community
Checking that the SKILL.md file is still in the repository.
Evaluate whether a Stripe dispute is economical to contest, then gather evidence and submit only when approved. Use for Stripe disputes, chargebacks, counter-disputes, evidence packages, or Stripe dispute IDs.
Install
npx vskill@latest install openclaudia/openclaudia-skills/stripe-disputeSKILL.md
Stripe Dispute Evaluation and Evidence
Evaluate the economics first. When contesting is justified and approved, build an evidence package and submit it to Stripe. Works for any SaaS using Stripe and a user database with login or usage logs.
When to Use This Skill
Use this skill when the user:
- Receives a Stripe chargeback notification and wants to fight it
- Provides a dispute ID (
du_*) and asks to "counter" / "rebut" / "fight" it - Asks how to gather evidence for a dispute marked
fraudulent,product_not_received,product_unacceptable, orsubscription_canceled - Wants an activity-log PDF showing a customer used their product
Required Environment Variables
Database safety: all queries are SELECT-only. Never let this skill issue UPDATE/DELETE/INSERT.
Inputs
The user provides any of:
- Stripe dispute ID (
du_xxxxx) — preferred - Customer email — skill will look up the dispute
- Charge ID (
ch_xxxxxorpy_xxxxx)
Decide whether to contest first
Accepting a dispute can be the correct outcome. Never submit evidence merely because a case exists.
- Verify live Stripe pricing. Under current US standard pricing, Stripe charges a $15
dispute-received fee automatically and another $15 dispute-countered fee when evidence is submitted. The received fee is sunk. The countered fee is returned only on a win. Use current live pricing if these amounts change.
- Query the merchant's live Stripe dispute history. For defended win rate, count only
closed disputes with evidence_details.submission_count > 0. Separate initial purchases from renewals and narrow to the same reason when the sample permits. Report wins, losses, open cases, and sample size. State when no exact-match precedent exists.
- Calculate break-even probability and expected value after the countered fee and
evidence-building labor. Do not invent a probability from case strength alone.
- Use a conservative 30% minimum win-probability threshold unless the user provides a
different threshold. If comparable data cannot support that threshold, recommend leaving the dispute unanswered. This avoids the countered fee, although the disputed amount and received fee remain lost.
- Obtain explicit user approval before building files, uploading evidence, or submitting.
Submission is final and fee-bearing.
Steps
1. Pull the dispute from Stripe
Extract: amount, reason, charge, evidence_details.due_by, evidence_details.submission_count, status.
If submission_count > 0 the dispute has already been countered — STOP and warn the user.
2. Pull the surrounding context
Prior undisputed payments on the same card are useful corroboration for fraudulent claims. Always count them, but do not treat them as proof by themselves.
3. Look up the customer in your app database
Adapt these queries to your schema. A useful evidence shape:
For product_not_received claims, check the user's self-reported cancel_reason. If they cancelled citing "Poor user experience" or anything admitting use, that field directly contradicts the claim. Quote it verbatim, but do not infer a win rate from one fact.
4. Download supporting documents
5. Capture your terms / cancellation policy as a PDF
Using Playwright (Node):
6. Generate the activity-log PDF
Build a concise, reviewable HTML activity log, then convert it to PDF with WeasyPrint:
HTML must include `<meta charset="UTF-8">` to avoid mangled characters in customer names/addresses.
A reviewable layout:
- Summary grid — Customer / Email / Internal user ID / Stripe customer ID / Account created / Plan / Subscription start / Cancellation timestamp (with delta from signup) / Self-reported cancel reason / Credits or units consumed / Purchase IP / Billing address / Payment method (last4, brand) / Stripe risk assessment / Referral source / Payment trigger
- Payment History table — every invoice with date, amount, status. Highlight the disputed row in red. Add a column "Disputed?" with explicit Yes/No.
- Checkout Attempts table — proves deliberate purchase intent (defeats
fraudulentby extension) - Items Created / Used table — names, types, timestamps. Concrete usage > adjectives.
- Login Activity Log — every session: timestamp, country, device. Consistency = same cardholder.
- Timeline Summary — single chronological narrative ending with "and then on $DATE the dispute was filed."
- Conclusion paragraph — quote the user's own
cancel_reasonagainst the stated dispute reason, if they contradict.
Prefer concrete numbers over adjectives. "29 login sessions, 3 named projects, 29,960 credits consumed" is more useful than "used extensively."
7. Show everything to the user before submitting
Display the rebuttal text, the evidence files, and your win-probability assessment. Wait for explicit user approval.
8. Upload evidence files to Stripe
Files go to `files.stripe.com`, NOT `api.stripe.com` — different host:
Returns {"id": "file_xxx", ...}. Capture the id — that's what you reference in evidence fields.
One file_id per dispute. Stripe rejects with 400 "That file is already attached to something else" if you try to reuse a file_id across disputes (especially for service_documentation). When fighting N disputes for the same customer, upload N copies of every shared PDF — same content, fresh file_id each time.
9. Submit evidence (one shot — submit=true is final)
Verify the response:
statusshould beunder_reviewevidence_details.has_evidenceshould betrueevidence_details.submission_countshould be1
Evidence Strategy by Dispute Reason
fraudulent
Goal: prove the cardholder made the purchase.
- Prior undisputed payments on the same card (strongest)
- OAuth login (Google/Apple) = identity-verified signup
- Consistent IP / country / device across sessions
- Multiple checkout attempts before purchase = real person deliberating
- Stripe's own risk assessment was
normal - Subscription still active and not cancelled
product_not_received
Goal: prove delivery + use.
- Login activity log
- Items created / actions taken inside the product
- The customer's own self-reported cancel reason, if they cancelled — quoted verbatim against the dispute claim
- Receipt and welcome email
product_unacceptable
Goal: show the product matched its description and the customer used it.
- Same as
product_not_receivedPLUS your terms-of-service language about quality / refund policy - Highlight that customer never opened a support ticket
subscription_canceled
Goal: prove the customer never cancelled (or cancelled after the renewal).
- Subscription object showing
cancel_at_period_end=falseat the renewal date - All login/use activity from after the renewal date
- Terms language stating annual renewals require explicit cancellation
Rebuttal Templates
uncategorized_text
[Customer name] created a [Product name] account on [date] via [auth method] and subscribed to [plan] ($[amount]/[interval]) using the same [card brand]. The first [N] payment(s) were never disputed. The customer actively used the service: [N] login sessions from [country] on [device], [N] items created ([list]), and [N] [units] consumed. The disputed charge is the [renewal/initial] payment on [date]. The subscription was [status] and remains [active/cancelled]. The customer never contacted support to cancel or request a refund. Our cancellation and refund policies are published at [TERMS_URL]. This is not a fraudulent transaction — it is a legitimate purchase from the cardholder who [made/has made] [N] other undisputed payments on this account.
cancellation_policy_disclosure
Our cancellation policy is disclosed at [TERMS_URL]. Subscribers may cancel at any time and retain access through the end of their billing cycle. This customer never cancelled.
This is the start of the file. Read the full SKILL.md on GitHub.
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/stripe-dispute/SKILL.md
Popularity
- Repository stars
- 426
- Repository forks
- 9
- 7-day trend score
- 30
Stars and forks are for the whole GitHub repository, not this skill alone. Refreshed 15 May 2026, 5 months ago.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/openclaudia/openclaudia-skills/stripe-dispute)