Payment Funnel Monitor

Monitor the payment last mile — initiate → success reconciliation → local method gaps — to recover conversion lost to failed checkouts. Use when payment succ...

Security Scan

An instruction-only skill whose requests and files match its payment-funnel analysis purpose and that does not require credentials, installs, or hidden network behavior.

Purpose & Capability

The name/description (payment last-mile monitoring, error-code analysis, local method recommendations) align with the SKILL.md content and the included reference material. It does not claim to connect directly to gateways; instead it expects merchant-provided context (platforms, date ranges, error-code samples, enabled methods), so the lack of required credentials or gateway access is proportionate.

Instruction Scope

Runtime instructions are constrained to reconciling attempts/success, grouping error codes, and recommending local methods. They reference only internal doc(s) (references/payment_codes_and_methods.md) and expected merchant data. There are no instructions to read unrelated system files, environment variables, or to call external endpoints.

Install Mechanism

No install spec and no code files beyond docs — the skill is instruction-only, so nothing is written to disk or downloaded. This is the lowest-risk install profile.

Credentials

The skill requests no environment variables, credentials, or config paths. Given it operates on merchant-supplied exports and descriptions, this is reasonable. If automated gateway access were desired, additional credentials would be expected, but none are requested here.

Persistence & Privilege

always:false and standard user-invocable/autonomous settings. The skill does not request persistent system presence or modify other skills. Autonomous invocation is allowed by default but does not combine with other red flags here.

Scan Findings in Context

[no_regex_findings] expected: Scanner found nothing to analyze because this is an instruction-only skill (no executable code). This is expected; review of SKILL.md and reference docs is the primary surface.

Assessment

This skill appears internally consistent and safe in that it does not request credentials or install code — it expects you to provide exports, error-code samples, and details about enabled payment methods. Before using it: (1) do not paste full card numbers or full PANs — provide anonymized/exported decline codes and tokenized IDs or aggregated counts instead; (2) if you need automatic gateway reconciliation, expect a future version to request gateway API keys or webhooks — only provide those to trusted code and rotate keys after use; (3) test the skill on a small, anonymized data set first to confirm outputs meet your needs; (4) if you see any later versions that add install steps, external endpoints, or requests for many unrelated credentials, treat that as a red flag.

You own pay step → authorization outcome: metrics, error code interpretation, and payment-method completion so fewer sessions die at the last click.

When to lean in

• Success rate < ~95% (define: successful charges / payment attempts or sessions reaching pay — align with merchant).
• Country anomalies (BR, DE, IN, etc.) vs global baseline.
• Timeout / decline ticket volume high.

Core workflow

1. Reconcile funnel — Initiated payment → submitted → succeeded vs failed vs abandoned (if data exists).
2. Error code analysis — Group by gateway code / decline reason; map to action (retry, 3DS, enable local rail, fraud rule).
3. Local method strategy — Per underperforming country, recommend wallets, bank transfer, installments, local cards as appropriate.

Gather context

1. Platform + gateways (Stripe, Adyen, PayPal, Shopify Payments, etc.).
2. Date range; overall and by country success rate.
3. Sample error codes or ticket themes.
4. Current payment methods enabled per market.

Read references/payment_codes_and_methods.md for code families and country method hints.

Mandatory success outputs (every full run)

1) Failure error code analysis

Structured block — table or bullet groups:

Error / decline family	Example codes (illustrative)	Likely cause	Action
Insufficient funds	issuer codes	Customer	Messaging; retry
3DS / SCA failed	authentication_required	Bank / UX	3DS flow, fallback
Timeout	gateway timeout	Latency / mobile	Timeout config, alternate method
Blocked / fraud	do_not_honor, risk	Rules / issuer	Review rules; local method
…	…	…	…

Use merchant’s actual codes when provided; otherwise use gateway-typical families and label as examples.

2) Recommended local payment methods (checklist)

Per priority country (or globally), output a checklist table:

Market	Recommended method	Role	Enable via (typical)
Brazil	Pix	Instant, high success	Stripe Pix, PSP
Brazil	Boleto (if applicable)	Unbanked	Local PSP
Germany	PayPal, SEPA, Klarna	Trust + bank debit	Gateway settings
…	…	…	…

At least three rows when multi-country; tie rows to observed underperformance or generic BR/DE if user named them.

KPI snapshot (include when possible)

Metric	Value	Target
Payment success rate	…	≥95%
Success by country (worst 3)	…	vs global
Timeout / unknown fail share	…	minimize

When NOT to use

• Shipping-only abandonment with no payment attempt data.
• Merchant payout settlement with no authorization funnel ask.

Split with other skills

• Abandoned checkout (full funnel) — use when drop is before pay; this skill is pay-step deep.
• Promo traffic stress — use when whole-site CVR drops under load.