Flèche Prêt Pro's loan and repayment records, reached under Ivorian authorization

A Flèche Prêt Pro account is keyed to a phone number, and behind that number sits the part worth integrating: the loan request an applicant submitted, the repayment schedule the lender computed at approval, and the identity material gathered at sign-up. The app describes itself as serving Côte d'Ivoire, with a loan band of 30,000 to 500,000 XOF and terms of 3 to 24 months. None of that ledger is on the handset. It is server state, and server state is reachable with the borrower's or the operator's authorization. That is the subject of this brief.

The honest read: this is a small originate-and-collect lender, so the value is the loan ledger and the schedule maths, not a treasury feed. A back office consolidating books across several such apps, a collections platform, or a credit file that needs the real instalment plan all want the same surface. We would build it on authorized traffic analysis of the app's own API, treating the borrower-facing consent path as the second lane for products that act on one user's records.

What the account stores

Each row below is something this specific app holds, not a generic lending checklist. Granularity matters here because the schedule and the fee accrual are where integrators usually get the maths wrong.

Authorized routes

Authorized interface integration and protocol analysis

The app talks to its own backend over HTTPS. Under the customer's authorization we observe and document that interface — auth, the loan and schedule endpoints, status transitions — and rebuild it as a clean client. Everything in the table above is reachable this way. Effort is moderate; durability is medium because endpoint shapes shift, so re-validation is part of how we maintain it. Access is arranged with you during onboarding against a consenting account or a sandbox, not requested as a hurdle before work begins.

Borrower-consented credential access

For a product that surfaces a single borrower's own loan and schedule, the borrower consents and we drive their authenticated session to read and export their records. Scope is that user's data only; durability is good per user; the consent capture and log are built in.

Native export as a cross-check

If the app emits any statement, PDF or CSV of the schedule, we parse it as a cheap secondary signal to validate the live integration. It is a supplement, never the spine.

A regulated account-information consent flow is not the route here. WAEMU's interoperable instant-payment system under BCEAO is live across the union, but consumer data-sharing on top of it is still groundwork-stage, so there is no general AIS scheme to consent through in Côte d'Ivoire today. In practice the build sits on the first route, with consented credential access layered on where a borrower-facing feature needs each user's own ledger, and any native export folded in as the validation lane. If BCEAO publishes a data-access framework, the integration is re-fitted onto it.

The package

You receive an OpenAPI specification covering the loan, schedule, status and identity surfaces actually present in this app. Alongside it: a protocol and auth-flow report tracing the phone-number plus OTP session and its token refresh; runnable source for the key endpoints in Python and Node.js, with the repayment-schedule and disbursement-status calls implemented and exercised; an automated test suite that asserts the instalment maths against the app's own worked example; interface documentation; and a data-retention and consent-logging note written against Ivorian law rather than a generic template. Each artefact maps to a surface above — there is nothing in the deliverable that this app does not expose.

A schedule pull, in practice

The shape below is illustrative and is confirmed against live responses during the build, not asserted as the app's current contract.

POST /api/v1/loan/schedule
Authorization: Bearer <session_token>        # phone-number + OTP session
Content-Type: application/json

{ "application_id": "FPP-... ", "include": ["instalments","fees"] }

200 OK
{
  "application_id": "FPP-...",
  "principal_xof": 30000,
  "apr_pct": 12.0,                 # within the 3.5–12% band the listing states
  "term_months": 4,
  "status": "disbursed",
  "instalments": [
    { "n": 1, "due_date": "2026-06-17", "amount_xof": 7795.89, "interest_xof": 295.89 }
    /* ... */
  ],
  "late_fee": { "daily_pct": 0.1, "cap_pct_of_overdue_principal": 18 }
}

# Error path we handle: 401 -> re-auth via OTP refresh; 409 -> application not yet approved (poll)

The interest figure ties back to the listing's own example — a 30,000 XOF loan at 12% over 120 days costing about 1,183.56 XOF — which tells us the maths is server-authoritative. We map the schedule endpoint instead of re-deriving interest on the client, so the integration stays correct if the lender retunes its rate band.

ARTCI and Law No. 2013-450

Personal data in Côte d'Ivoire is governed by Law No. 2013-450 of 19 June 2013, with ARTCI designated as the data-protection authority; processing must be declared to ARTCI in advance and sensitive processing pre-authorized, per the law's text and ARTCI's own guidance. ARTCI has been pointed about this app category specifically: in an August 2023 communiqué and a follow-up, it warned about online-lending apps collecting personal data without authorization and naming reports of contact-list misuse. We treat that as the operating constraint, not a footnote. Access is authorized or borrower-consented; consent scope is recorded with an expiry and a revocation path; the contact and identity captures are minimized to what the integration needs, which is also what the law's finality and proportionality principles demand. An NDA covers the engagement where the client requires one.

Engineering realities we plan around

Three things about this app shape how we build, all handled on our side. The repayment maths is computed at approval and lives on the server, so we model the schedule endpoint as the source of truth and never re-derive interest on the device; that keeps the feed accurate when the 3.5%–12% band is retuned. Sign-up is phone-number-first with an OTP, which binds the session to a device or SIM, so we design the token-refresh and re-auth path around that and a long-running sync does not silently expire — the test account or sandbox needed for that is arranged with you during onboarding. The app gathers contact-list and identity material at KYC, so per-field minimization is designed into the extraction from the start, both to satisfy ARTCI's proportionality principle and to keep the resulting data set defensible. When the front end is restyled, a re-validation pass against the documented endpoints is part of maintenance.

Where it gets used

Three end-to-end shapes that come up with lenders of this size:

• A multi-lender back office pulls the loan and schedule ledger nightly to consolidate books and flag a borrower who is live on several apps at once.
• A collections platform subscribes to the disbursement-status surface, then drives dunning from the real instalment due dates and the 0.1%-per-day accrual rather than an estimate.
• A credit-file or AML tool reads the KYC and identity fields through a borrower-consented session to build a verifiable file with a recorded consent trail.

Interface evidence

Screenshots from the listing, used only to confirm the surfaces named above. Select to enlarge.

Other Ivorian lenders in the same data shape

Same-category apps in Côte d'Ivoire that hold a comparable loan-and-schedule ledger. Listed for ecosystem context, since a unified integration usually has to span several of them.

• Prêt Facile Côte d'Ivoire — short-term loan applications and repayment tracking for the Ivorian market.
• CI Money — instant loans in the 30,000–500,000 FCFA band with 3-to-24-month terms, an account holding application and schedule data.
• Joli Prêt — small short-tenor loans with day-rate interest and a per-borrower schedule.
• Prêt Shell — fast cash-loan service holding application status and repayment records.
• SmartLoan Côte d'Ivoire — a domestic online-lending platform with borrower accounts and disbursement state.
• Credit Rapide — quick-credit app with per-user application and balance data.
• Pretci — Ivorian online-loan service holding loan and repayment history.
• Xof-Royaume — fast XOF loans with an account-held schedule and arrears state.

How this was built

Checked in May 2026: the app's own Play Store description for amounts, term, APR band and the late-fee rule; ARTCI's data-protection pages and its communiqué on online-lending apps; the text of Law No. 2013-450; and regional reporting on BCEAO's PI-SPI interoperable payment system to confirm the open-banking status. Primary sources opened:

• ARTCI — personal data protection (regulator)
• ARTCI communiqué — online-lending mobile apps
• Law No. 2013-450 of 19 June 2013 (English text)
• WAEMU's interoperable payment system — background

OpenBanking Studio · integration desk, May 2026

Questions integrators ask about Flèche Prêt Pro

Flèche Prêt Pro lists loans of 30,000 to 500,000 XOF — does the integration expose the full repayment schedule or only the outstanding balance?

Both. The schedule is computed server-side at approval, not on the handset, so we map every instalment, its due date, the principal and interest split, and the late-fee accrual the listing describes (0.1% per day, capped at 18% of overdue principal). The outstanding balance is then a derived view over that ledger rather than the only field you can read.

Which Ivorian body governs the data this app collects, and how does that shape the build?

ARTCI, acting as the personal-data authority under Law No. 2013-450 of 19 June 2013. We work from authorized or borrower-consented access, keep a consent and processing log, and scope extraction to the fields the integration needs — the contact list and identity captures stay minimized because the law's finality and proportionality principles require it.

Is there a regulated open-banking consent route for a Côte d'Ivoire lender like this one?

Not yet as a general scheme. BCEAO's interoperable instant-payment rails (PI-SPI) are operational across the WAEMU, but consumer-facing data sharing is still groundwork-stage, so for this app we use authorized interface integration and protocol analysis rather than a regulated AIS consent flow. If BCEAO publishes a data-access framework we re-fit the integration onto it.

The app approves loans asynchronously — can the integration track status changes without scraping the screen?

Yes. Approval transitions are server-side state keyed to the application id, so we model the status surface directly and run a poll-plus-callback hybrid against it. The screen is treated as a fallback signal, not the source of truth, which keeps a disbursement feed accurate even when the front end is restyled.

Starting the build

Source delivery for Flèche Prêt Pro starts at $300 — you receive the runnable endpoints, the OpenAPI spec, tests and the interface documentation, and you pay only after delivery once the build satisfies you. The alternative is our hosted endpoints: you call them and pay per call, with nothing upfront. Either way the cycle is one to two weeks, and all you supply is the app name and what you want from its data; access and the ARTCI-aligned compliance posture are arranged with you as part of the engagement. Tell us what you need at /contact.html and we will scope it.