Back to blog
Standards & methodology2026-07-157 min read

AAOIFI Screening API: Methodology & Data Fields

A developer reference for Akinda's AAOIFI-based screen: the three-phase methodology, the 30% market-cap ratio thresholds, the halal_status verdict values, and exactly which fields each REST endpoint returns.

Editorial cover on the Akinda brand gradient titled "AAOIFI Screening API: Methodology & Data Fields", with four correct stat tiles: "3-phase screen", "30% ratio threshold", "verdict: halal_status", and a real AAPL tile "debt_ratio_perc 2.99%".

When you embed a Shariah verdict in your product, your users — and eventually their Shariah board — will ask the obvious question: how was this decided? An AAOIFI screening API answers that by making the methodology explicit and returning the data behind every verdict. This guide breaks down the three-phase methodology the Akinda Halal API implements, the thresholds it applies, the real fields each endpoint returns, and how AAOIFI-based screening differs from the index methodologies you may have seen elsewhere — so you can build features your compliance team can actually audit.

Quick answer

Akinda runs an AAOIFI-based, three-phase screen: a business-activity test that removes impermissible revenue, two financial-ratio tests (debt and liquidity, each against a 30% market-cap threshold), and a purification step for incidental impermissible income. The REST endpoints return a halal_status verdict plus supporting ratio fields; purification is a separate calculation exposed through the Akinda MCP server. What you get back depends on your tier — the free tier returns the verdict, paid tiers add the ratios and figures.

What is AAOIFI screening?

AAOIFI — the Accounting and Auditing Organization for Islamic Financial Institutions — sets the most widely used standards in Islamic finance. Its Shari'ah Standard No. 21 defines how to screen shares for compliance: first by what the company does, then by its financial ratios. An AAOIFI screening API encodes those rules so your software can apply them consistently to any ticker, at scale. Akinda covers US tickers broadly, with UK (.L) and Canada (.TO) listings on higher tiers — work that would be impractical to do by hand, but only valuable if the methodology is sound and transparent.

The three-phase methodology

Akinda runs the screen in three phases. A stock must clear all three before the API returns halal_status: "HALAL"; fail any one and it returns NOT HALAL (or DOUBTFUL where the evidence is mixed).

1. Business-activity screen

First, the company's core business must be permissible. Companies whose primary revenue comes from prohibited activities fail here regardless of how healthy the balance sheet looks. The classic excluded sectors are:

  • Conventional finance — interest-based banking, insurance, and lending — because it runs on riba.
  • Alcohol, tobacco, and pork production and distribution.
  • Gambling and betting (maysir).
  • Adult entertainment and other clearly impermissible content.
  • Weapons tied to non-permissible use.

The screen also accounts for incidental income: a company that mostly does something permissible but earns a small share of revenue from a prohibited source may still pass, provided that share stays under AAOIFI's non-compliant-income tolerance of 5% of revenue. Akinda quantifies this as non_compliant_revenue_perc — for Apple it is 0 — so it's a revenue-composition test, not a keyword filter.

2. Financial-ratio screen

Companies that pass the business screen then face two financial tests, each measured against 30% of the company's market capitalisation (Standard 21 uses market cap, not total assets, as the base):

  • debt_ratio_perc — interest-bearing debt against the threshold. A company financed largely by conventional, interest-bearing borrowing fails.
  • liquidity_ratio_perc — interest-bearing investments and liquid assets against the threshold, capturing riba exposure on the asset side.

Both must stay within the limit. For Apple the API returns debt_ratio_perc 2.99 and liquidity_ratio_perc 1.66 — comfortably clear of the 30% line, which is why AAPL screens HALAL.

3. Purification

Even a HALAL company may earn a small amount of incidental impermissible income — for example, interest on cash balances. Purification quantifies the portion of dividends attributable to that income so a shareholder can give it to charity rather than keep it. In Akinda this is a calculation rather than a static report field: the Akinda MCP server exposes an akinda_calculate_purification tool that returns the amount to cleanse for a given holding, so your app can turn compliance into an actionable number instead of leaving users to estimate.

Endpoints and the data fields you get back

Rather than a bare pass/fail, Akinda surfaces the data behind the verdict so it's auditable. There are three REST endpoints under https://b2b-api.akinda.io/api/v1, and the ticker is a path parameter (not a ?symbol= query string). Authenticate with your key as either an X-API-Key header or an apikey query parameter — there is no Authorization: Bearer scheme.

cURL

curl "https://b2b-api.akinda.io/api/v1/compliance/AAPL" \
  -H "X-API-Key: YOUR_API_KEY"

The three endpoints, from lightest to fullest:

  • GET /compliance/{ticker} — the verdict only. This is the free Basic tier (200 calls/day) and is built for speed.
  • GET /basic-report/{ticker} — the verdict plus the three AAOIFI ratios above (debt_ratio_perc, liquidity_ratio_perc, non_compliant_revenue_perc). Available from Personal Standard.
  • GET /full-report/{ticker} — everything in the basic report plus a halal_rating, AI-extracted figures from SEC filings, and their dollar counterparts. Available from Personal Ultimate. The Akinda ranking pillars, such as ranking_score, are reserved for Business Ultimate and Enterprise.

A compliance response is deliberately minimal:

JSON

{
  "company_name": "Apple Inc.",
  "halal_status": "HALAL",
  "ticker": "AAPL"
}

A basic report adds the three ratio fields:

JSON

{
  "company_name": "Apple Inc.",
  "ticker": "AAPL",
  "halal_status": "HALAL",
  "debt_ratio_perc": 2.99,
  "liquidity_ratio_perc": 1.66,
  "non_compliant_revenue_perc": 0
}

The full report layers on a halal_rating, the dollar counterpart behind each ratio, and the AI-extracted totals from the filings:

JSON

{
  "company_name": "Apple Inc.",
  "ticker": "AAPL",
  "halal_status": "HALAL",
  "halal_rating": 10,
  "debt_ratio_perc": 2.99,
  "debt_ratio_dollar": 98657000000,
  "liquidity_ratio_perc": 1.66,
  "non_compliant_revenue_perc": 0,
  "reporting_currency": "USD"
}

The ranking pillars — ranking_score and the profitability, growth, and momentum scores — are gated to Business Ultimate and Enterprise, so they won't appear on a Personal Ultimate full report.

Current verdict: Halal

Apple Inc. (AAPL) screens halal_status HALAL — debt_ratio_perc 2.99 and liquidity_ratio_perc 1.66, both far inside the 30% line, with non_compliant_revenue_perc 0. The full report adds a halal_rating of 10.

Beyond HALAL and NOT HALAL, a few more verdict values are worth handling. halal_status can also come back as DOUBTFUL when the evidence is mixed, or as ERROR_DATA / INCOMPLETE_DATA when the underlying filings don't give the screen enough to work with — handle those explicitly rather than treating a missing verdict as a pass. The exact field list per endpoint is in the developer docs.

Why the denominator matters

The choice of denominator is where screening methodologies quietly differ, and it matters for your users. Measuring ratios against a single day's market cap makes verdicts jumpy: a sharp price move can flip a stock in or out of compliance overnight, even though nothing about the business changed. Averaging market cap over a trailing window smooths that noise, producing more stable, defensible verdicts and fewer whipsaw alerts.

When you evaluate any AAOIFI screening API, check what denominator it uses, because two APIs applying "the same" 30% rule to different denominators can legitimately disagree on the same stock. A published denominator is a sign the provider has thought about verdict stability, not just the headline threshold.

How AAOIFI compares to index methodologies

AAOIFI is not the only Shariah screening framework. Index providers — Dow Jones Islamic Market, S&P Shariah, FTSE, and MSCI Islamic — each publish their own rules, and they differ in two ways that explain why screeners sometimes disagree on the same company:

  • The denominator. Some methodologies measure ratios against market capitalisation (as AAOIFI Standard 21 does); others use total assets. A company can pass under one and fail under the other purely because of the base.
  • The thresholds. The exact percentage limits for debt and interest-bearing assets vary between frameworks, so a borderline company can land on either side depending on which ruleset you apply.

None of this makes one framework "right" and another "wrong" — they encode legitimate scholarly differences. What matters for your product is picking a transparent, consistently applied methodology and being clear about which one your verdicts reflect. For a deeper side-by-side, see AAOIFI vs IFSB vs S&P Shariah.

Governance and auditability

A methodology is only trustworthy if it's transparent and governed. The lack of standardised Shariah interpretation across jurisdictions is one of the industry's biggest challenges, which is exactly why published rules matter. Akinda's screen follows AAOIFI Shari'ah Standard No. 21, with the ratio formulas and business tests documented so your integrators — and their own Shariah boards — can audit the math rather than trust a black box. Edge cases such as mergers, purification, and partial-period reporting are reviewed under qualified Shariah governance.

That combination — published rules plus scholarly oversight — is what lets you defend a verdict to a regulator, a user, or a Shariah committee. Treat the output as data-driven signals, not a fatwa: pair it with your own Shariah governance for final rulings. If you're wiring this into a product, adding Shariah compliance to your app walks through the integration end to end.

Frequently asked questions

Is Akinda's screening AAOIFI-certified?

The methodology is AAOIFI-based — it follows Shari'ah Standard No. 21 and is documented so it can be audited against that standard. As with any automated screen, treat the output as data-driven signals rather than a fatwa, and keep your own Shariah governance in the loop for final rulings.

What financial ratios does the API check?

Two, after the business-activity screen: a debt ratio (debt_ratio_perc, interest-bearing debt) and a liquidity ratio (liquidity_ratio_perc, interest-bearing investments and liquid assets), each measured against 30% of market capitalisation. A company must clear the business screen and both ratios to return HALAL. These ratio fields ship on the basic-report endpoint (Personal Standard and above).

Does the API return purification amounts?

Purification is a calculation, not a static report field. The Akinda MCP server exposes an akinda_calculate_purification tool that returns how much of a holding's dividend income to give to charity. See connecting Akinda to Claude via MCP, and confirm the treatment with your Shariah board before displaying figures.

Why do two Shariah screeners sometimes disagree?

Because they can use different denominators, thresholds, or source data. Two APIs applying a 30% rule will still differ if one measures against a single-day market cap and another against a trailing average, or if one uses market cap and another total assets. That's why a transparent, published methodology matters: it lets you see exactly why a verdict came out the way it did.

Which endpoint should I call?

Use /compliance/{ticker} when you only need the verdict — it's on the free tier at 200 calls/day. Step up to /basic-report/{ticker} for the three AAOIFI ratios, or /full-report/{ticker} for the halal_rating, SEC-derived figures, and dollar counterparts.

Verify it yourself via the Akinda API

Fire a live /full-report/<ticker> call from the playground using your own API key — see the compliance ratios, AAOIFI screen verdict, and source-breakdown fields the methodology produces.

Related reading