
sumsub-integrate-id-connect
End-to-end recipe for integrating Sumsub ID Connect — the OIDC-based "Verify with Sumsub ID" flow that lets a user share previously-verified identity claims with your app. TRIGGER when the user asks to "integrate Sumsub ID Connect", "add Verify-with-Sumsub-ID button", "reuse Sumsub KYC via OIDC", "exchange the Sumsub authorization code", "mint a Sumsub ID share token", or asks how to wire up @sumsub/id-connect, the OIDC code → accesstoken exchange (/api/snsId/oauth/token), the share-token / partner-consent flow, or the share-link redirect. Covers preconditions checklist, automated preflight, frontend button (library button + raw redirect alternative), backend code-for-token exchange, share-token minting, partner-consent shareLink flow, and applicant materialisation through /resources/api/reusableIdentity/reuse. SKIP for the standalone WebSDK / KYC widget (use sumsub-integrate-websdk), or for plain Reusable-KYC-via-API where the donor and recipient are both your own tenants w
- Installs
- 981
- Last 24h
- +16
- Trending rank
- #6692
- Security
- Not audited
Install
Works with every agent the skills CLI supports.
npx skills add https://github.com/sumsub/agent-skills --skill sumsub-integrate-id-connect- 01
Install
Run the command in your project. The CLI asks which agents to install for.
- 02
Describe the task
Ask your agent for something this skill covers. It loads sumsub-integrate-id-connect automatically when relevant.
- 03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.
SKILL.md
The instructions your agent reads when the skill is triggered.
Sumsub — ID Connect integration
Embed the "Verify with Sumsub ID" OIDC flow end-to-end — preconditions → preflight → button → backend exchange → consent → applicant in your workspace.
⚠️ Sandbox tokens only
Do not use a production App Token / secret while wiring this up. Insist on
a sandbox pair from
https://cockpit.sumsub.com/checkus/home?sbx=true — Connect Sumsub to
your AI agent -> Build & configure -> Generate token. Token + secret
are revealed once at creation; copy both before closing the dialog. The preflight enforces this
with an sbx: prefix check.
Deeper auth mechanics: sumsub-api-auth.
⛔ Order of operations — do NOT scaffold an app first
This is the single most important rule of this skill. Write no frontend or backend code, create no demo project, scaffold nothing until BOTH gates below are green. The client-side Sumsub setup (Stage 0a) is manual and cannot be done by you — jumping to code before it exists produces an app that can't authenticate, can't redirect, and can't reuse. Always follow this sequence:
- Confirm Stage 0a preconditions with the user — one by one. Do not assume any item is done. For each, ask the user whether it exists; if it doesn't, stop and hand them the exact manual step and where to do it (the support email template, the Dev Space → OIDC Settings registration, the App Token minting page, the level creation hand-off). These are Sumsub-dashboard / support actions only — there is no public API for them, so you cannot do them for the user. Wait until every item is confirmed before moving on.
- Run Stage 0b preflight and require exit 0. Only after the preconditions
are confirmed, run
scripts/preflight.sh. Resolve every FAIL (loop back to Stage 0a as needed). Do not proceed on any FAIL. - Only now build. With both gates green, proceed to Stage 1 and scaffold the frontend/backend.
If the user says "just build the demo" before the gates are green, briefly explain why it won't work yet and walk them through the missing precondition first. The code is the last step, not the first.
What ID Connect is (and isn't)
Sumsub ID Connect lets a business request specific verified identity claims — age, name, nationality, email, country of residence, humanness — from a Sumsub ID account belonging to the end-user. Sumsub hosts the entire user-facing flow: sign-in, account creation (if needed), email OTP, any missing verification steps, and a per-recipient consent screen. Your app receives the result as a signed OIDC token plus, optionally, an applicant materialised in your workspace through Reusable KYC.
This is not the WebSDK. The WebSDK runs a full KYC capture inside your page; ID Connect delegates the whole UX to Sumsub.
The lifecycle in one picture
┌──────────────────────────────┐
│ Stage 0a Preconditions check │ manual setup on Sumsub side
│ Stage 0b scripts/preflight.sh │ automated validation
└──────────────┬───────────────┘
│ clientId, levelName, allowlist
┌──────────────▼───────────────┐
│ Stage 1: Frontend "Verify…" │ authorize redirect with PKCE
│ button │
└──────────────┬───────────────┘
│ onSuccess({ code, state })
┌──────────────▼───────────────┐
│ Stage 2: code → access_token │ POST id.sumsub.com/api/snsId/oauth/token
│ (client_secret_basic) │
└──────────────┬───────────────┘
│ access_token
┌──────────────▼───────────────┐
│ Stage 3a: → share token │ POST /resources/accessTokens/sumsubIdShareToken
└──────────────┬───────────────┘
│ sharingAllowed?
┌───────┴────────┐
│ false │
▼ ▼ true
┌──────────────────┐ ┌──────────────────────────────┐
│ Stage 3b: │ │ Stage 4: reuse/preview + │
│ shareLink │ │ reuse/commit │
│ redirect + retry │ │ → applicant in your workspace │
└────────┬─────────┘ └──────────────────────────────┘
│
└──────► back to Stage 3a (retry — now sharingAllowed: true)
Endpoints
| METHOD | URL | Auth | Stage |
|---|---|---|---|
| (browser redirect) | https://id.sumsub.com/api/snsId/oauth/authorize | none | 1 |
| GET | https://id.sumsub.com/.well-known/openid-configuration | none | 0b |
| GET | https://id.sumsub.com/.well-known/jwks.json | none | (id_token validation) |
| POST | https://id.sumsub.com/api/snsId/oauth/token | client_secret_basic | 2 |
| POST | https://api.sumsub.com/resources/accessTokens/sumsubIdShareToken | App Token + HMAC | 3a |
| POST | https://api.sumsub.com/resources/snsId/oauth/shareLink | App Token + HMAC + Bearer | 3b |
| GET | https://api.sumsub.com/resources/api/reusableIdentity/reuse/preview | App Token + HMAC | 4 |
| POST | https://api.sumsub.com/resources/api/reusableIdentity/reuse | App Token + HMAC | 4 |
The "Enable Sumsub ID Connect" toggle, OIDC client registration (client_secret + redirect_uri allowlist), and sharing-partner registration are Sumsub dashboard UI only — no public-API equivalent.
Stage 0a — Preconditions checklist
Before any code or preflight, the following must already exist on the Sumsub side. Each item maps to a check below or to a runtime failure if missed.
-
Sumsub support has enabled ID Connect for your workspace. This is not a self-serve toggle — until support flips it, no OIDC client can be created and the preflight will report
connect_token: ID Connect not enabled for this workspace. Email your CSM orsupport@sumsub.comfirst and wait for confirmation. Template:``` Subject: Enable Sumsub ID Connect for clientId <YOUR_CLIENT_ID> (sandbox) Hi Sumsub team, Please enable Sumsub ID Connect for our workspace. Environment: sandbox clientId: <YOUR_CLIENT_ID> (Dashboard top-left) Intended scopes: openid, share, name (adjust as needed) Intended recipients: self (or list partner clientIds) Use case: <one-line description of where the button will live> Once enabled we'll register the OIDC client + redirect URIs in Dev Space → OIDC Settings. Thanks ``` -
App Token + secret (sandbox). Minted at https://cockpit.sumsub.com/checkus/home?sbx=true via Connect Sumsub to your AI agent -> Build & configure -> Generate token. Both values shown once at creation. → exported as
SUMSUB_APP_TOKEN/SUMSUB_SECRET_KEY. -
OIDC client registered at Dashboard → Dev Space → OIDC Settings. Capture all three: -
client_id(public — goes in the browser, used asforClientIdfor the share token in the common self-recipient case) -client_secret(shown ONCE at creation; if you lose it you must regenerate, which invalidates existing integrations). Required for the Stage 2 code→token exchange — keep it server-side only. -redirect_uriallowlist populated for BOTH callbacks (Sumsub matches byte-for-byte — trailing slash, casing, port, query all matter): - your authorize callback (e.g.https://yourapp.com/auth/callback) - your shareLink callback (e.g.https://yourapp.com/share-complete) Sumsub redirects the browser back to these URLs and requires public HTTPS — a barehttp://localhost:3000won't work for local testing. To develop on your machine, expose your local server with a public HTTPS tunnel (ngrok) and allowlist that origin — see Local development with ngrok below. -
At least one verification level exists in this workspace. Capture
levelName. If none, hand off tosumsub-create-levelfirst — Stage 4 can't materialise an applicant without one. -
At least one Sumsub ID account with verified data for end-to-end testing. An email-only account (no completed KYC) makes Stage 4 fail with errorCode
10521 reusable-kyc-inactive-sumsub-id-account— see the error table at the end of Stage 4.
Stage 0b — Preflight
scripts/preflight.sh machine-validates the
preconditions that can be probed without minting OIDC codes or creating
applicants. It does not mutate anything.
SUMSUB_APP_TOKEN=sbx:... SUMSUB_SECRET_KEY=... \
bash scripts/preflight.sh
Checks (each independent — one failing doesn't abort the others):
| Check | What it proves |
|---|---|
connect_token | App-Token HMAC signing is accepted and ID Connect is enabled for the workspace — probes the endpoint with a known-bad code and expects 401 Invalid code (a 404 invalid clientId means ID Connect isn't enabled). This is a connectivity/entitlement probe, not the Stage 2 exchange route. |
oidc_discovery | https://id.sumsub.com/.well-known/openid-configuration returns a valid OIDC discovery doc |
levels | At least one verification level exists in this workspace |
Exit code: 0 on all-PASS / PASS+WARN, 10 on any FAIL. Resolve all FAILs
before Stage 1.
Local development with ngrok
ID Connect can't be tested against http://localhost. Sumsub redirects the
browser back to your redirect_uri after the authorize flow (Stage 1) and
after the shareLink consent flow (Stage 3b), and it requires the URL to be
public HTTPS and present in the OIDC client's redirect_uri allowlist
byte-for-byte. A bare localhost URL is neither public nor HTTPS, so the
redirect fails. For local development, put a public HTTPS tunnel in front of
your local server with ngrok and allowlist that origin.
-
Start your backend locally (the example listens on
PORT=3000):PORT=3000 node examples/express-callback.js -
Open a tunnel to that port. Prefer a reserved (static) domain so the URL survives restarts — a free random ngrok URL changes on every launch and forces you to re-edit the allowlist each time:
ngrok http 3000 --domain=your-reserved-name.ngrok-free.app # or, ephemeral URL (changes each restart): ngrok http 3000ngrok prints a forwarding URL, e.g.
https://your-reserved-name.ngrok-free.app. -
Allowlist BOTH callbacks at the ngrok origin in Dashboard → Dev Space → OIDC Settings (sandbox workspace) — byte-for-byte, including trailing slash:
https://your-reserved-name.ngrok-free.app/ https://your-reserved-name.ngrok-free.app/share-complete -
Point the app at the public URL, not localhost. The browser's
redirect_uriand the server-side shareLinkredirectUrimust derive from the same public origin, so setPUBLIC_BASE_URLand restart:PUBLIC_BASE_URL=https://your-reserved-name.ngrok-free.app \ PORT=3000 node examples/express-callback.jsThen open the ngrok URL (not
localhost:3000) in the browser so theredirect_uriit sends matches the allowlist.
⚠️ Re-register on every URL change. If you use an ephemeral ngrok URL, the allowlist entry (and
PUBLIC_BASE_URL) must be updated each time ngrok restarts. A reserved domain avoids this churn.ngrok's free interstitial ("You are about to visit…") only affects API/XHR calls, not top-level browser navigation, so it doesn't block the OIDC redirects. A reserved domain or paid plan removes it entirely.
Stage 1 — Frontend: the button
⛔ Gate. Do not start this stage until Stage 0a preconditions are all confirmed with the user AND Stage 0b preflight exits 0. If you haven't done both, go back — see "Order of operations" above. This is the first stage where you write code; everything before it is setup you must verify first.
Architecture — what's yours vs Sumsub's
Sumsub provides: the OIDC service (id.sumsub.com), the HMAC API
(api.sumsub.com), and the optional @sumsub/id-connect JS library that
mounts the button.
You write both halves of the integration:
- Frontend (this stage) — page with the button. Generates PKCE, kicks
off the OIDC redirect to
id.sumsub.com, receives?code=in the callback, POSTs{ code, codeVerifier }to your own backend. - Backend (Stages 2-4) — HTTP endpoint that your frontend POSTs to.
Signs Sumsub API calls with your App Token + secret (which must NEVER
reach the browser), orchestrates the chain
code → access_token → share token → reuse, and returns the materialised applicant.
The two halves are coupled by one route name — POST /api/sumsub/id-connect/exchange
in the examples; rename to fit your routing. Reference pair (matched):
examples/oidc-button.html (frontend) +
examples/express-callback.js (backend).
The backend is a thin orchestrator (~180 lines of Node/Express); port it to
your stack of choice — the wire format is identical regardless of language.
⚠️
onSuccessis not a verification signal. It only confirms the user finished the OIDC consent step and Sumsub issued an authorization code. The actual verification verdict comes from Stage 4 —applicant.review.reviewStatusin the/reuseresponse (immediate, if your recipient level runs no additional checks) or theapplicantReviewedwebhook (authoritative, for any level that runs post-reuse checks). Never grant access or unlock features based ononSuccessfiring.
Three variants — use Method 1 (createButton) by default; the others
exist for design-system or no-bundler scenarios.
Method 1 — pre-built button via @sumsub/id-connect (recommended)
import { createButton } from '@sumsub/id-connect';
// PKCE helpers (S256 — full versions in examples/oidc-button.html).
const b64url = (buf) => btoa(String.fromCharCode(...new Uint8Array(buf)))
.replace(/\+/g, '-').replace(/\//g, '_').replace(/=+$/, '');
const sha256 = (s) => crypto.subtle.digest('SHA-256', new TextEncoder().encode(s));
// Generate the PKCE pair yourself, store the verifier under YOUR key — Stage 2
// on the backend needs that verifier to exchange the code for an access token.
const verifier = b64url(crypto.getRandomValues(new Uint8Array(32)));
const challenge = b64url(await sha256(verifier));
sessionStorage.setItem('myapp.pkce.verifier', verifier);
createButton({
clientId: '<your-client-id>',
permissions: ['openid', 'share', 'name'],
container: document.getElementById('button-container'),
codeChallenge: challenge, // library uses YOUR challenge — skips its own PKCE gen
loginHint: 'user@example.com', // optional, prefills the email field
onSuccess: async ({ code, state }) => {
const v = sessionStorage.getItem('myapp.pkce.verifier');
await fetch('/api/sumsub/id-connect/exchange', {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({ code, codeVerifier: v, redirectUri: location.href }),
});
},
onError: (err) => console.error('sumsub-id-connect error', err),
});
Why generate PKCE yourself: the Stage 2 token exchange needs the verifier.
If you let the library generate it, you depend on the library's internal
sessionStorage key (undocumented per minor version) to read it back in
onSuccess. Passing your own codeChallenge keeps the verifier under a key
you own. Same pattern works for Methods 2 and 3.
Runnable example: examples/oidc-button.html.
Method 2 — your trigger + library modal
Same library, you own the trigger element; library opens the flow in an
overlay via openModal({...}). Use when your design system has its own
buttons.
Runnable example: examples/oidc-modal.html.
Method 3 — raw OIDC redirect (no library)
Direct navigation to the OIDC authorize endpoint, no Sumsub script on the page. Reach for this when CSP forbids third-party scripts, you're in an SSR-only environment, or you maintain a generic OIDC client across multiple providers.
GET https://id.sumsub.com/api/snsId/oauth/authorize
?client_id=<your-client-id>
&response_type=code
&redirect_uri=<allowlisted URI>
&scope=openid+share+name
&state=<random CSRF nonce>
&code_challenge=<base64url(SHA256(code_verifier))>
&code_challenge_method=S256
&login_hint=<optional email>
You generate the code_verifier (43–128 chars, URL-safe), derive
code_challenge = base64url(SHA256(verifier)), store the verifier
(sessionStorage), and forward it to your backend after the callback.
Sumsub only accepts code_challenge_method=S256 (plain is rejected) and
the redirect_uri must match the allowlist byte-for-byte.
Runnable example: examples/oidc-redirect.html.
Choosing scopes
Default ['openid', 'share', 'name']. Add email, profile, or
offline_access (refresh token) only when downstream logic actually reads
them — extra scopes trigger extra checks in the user-facing flow. Full table:
references/scopes-and-claims.md.
Stage 2 — Backend: code → access_token
Exchange the authorization code on the OIDC token endpoint. The code came from
the browser authorize flow, so its counterpart is the standard OIDC token
endpoint — authenticated with the OIDC client_secret (client_secret_basic),
not HMAC.
POST https://id.sumsub.com/api/snsId/oauth/token
Content-Type: application/x-www-form-urlencoded
Authorization: Basic base64(<client_id>:<client_secret>)
grant_type=authorization_code
&code=<the code your frontend got>
&redirect_uri=<exact-match URI from Stage 1 — byte-for-byte>
&code_verifier=<PKCE verifier matched to the code_challenge from Stage 1>
Response (24-hour access_token):
{
"access_token": "snd-id-con-a-...",
"refresh_token": "snd-id-con-r-...", // only with `offline_access`
"token_type": "Bearer",
"expires_in": 86400,
"id_token": "eyJhbGc..." // only with `openid` — RS256-signed
}
Refresh later via grant_type: "refresh_token"; you cannot request more
scopes on refresh than were granted at consent time.
The
client_secretstays server-side — never in the browser bundle. The exchange isclient_secret_basic/client_secret_post; do not HMAC-sign it. (The App-Token + HMAC auth is still used in Stage 3a/4 for the share token and reuse calls.)
Gotchas
Invalid code401 has three common causes: PKCE mismatch (verifier doesn't hash to the challenge sent at authorize), code already used (single-use; refreshing the callback page double-fires), or code expired (short TTL — exchange immediately).redirect_urimust match Stage 1 byte-for-byte — trailing slash, casing, port and query all count, or the exchange fails.id_tokenvalidation — use a mature library (jose,pyjwt,jjwt) against thejwks_urifrom.well-known/openid-configuration. Never trust unverified JWT contents.
Stage 3 — Backend: access_token → share token + consent
Mint a share token scoped to a recipient (forClientId). Always implement
the consent fallback (Stage 3b) — sharingAllowed: false is returned
whenever the user hasn't yet authorised sharing with this specific
forClientId. In the common self-recipient case both consents are usually
granted in one click on the Stage 1 consent screen, but production code that
only handles sharingAllowed: true breaks the first time a user re-enters
the flow or a recipient configuration differs.
Stage 3a — mint share token
POST https://api.sumsub.com/resources/accessTokens/sumsubIdShareToken
{
"sumsubIdConnectToken": "<access_token from Stage 2>",
"forClientId": "<recipient clientId>",
"ttlInSecs": 1800
}
forClientId:
- Self-recipient (common case): your own workspace
clientId, visible in the Dashboard top-left. - Third-party recipient: the partner's
clientId(they must be pre-registered as a partner in the Dashboard).
Response:
{
"token": "_act-snsId-...",
"forClientId": "<echoed>",
"sharingAllowed": true // or false → continue to Stage 3b
}
Stage 3b — partner-consent shareLink
If sharingAllowed: false, mint a consent link and redirect the user:
POST https://api.sumsub.com/resources/snsId/oauth/shareLink
Authorization: Bearer <access_token from Stage 2>
X-App-Token + X-App-Access-Ts + X-App-Access-Sig
{
"redirectUri": "<your shareLink callback URL — must be in the allowlist>",
"forClientId": "<same recipient clientId>",
"displayMode": "page",
"state": "<csrf-or-correlation-id>"
}
Response: { "link": "https://id.sumsub.com/share/<id>" } — expires in 5
minutes.
Persist the access_token server-side keyed by state (in-memory Map is OK
for demos; use Redis / encrypted cookie / session table for prod). Redirect
the user to the link. After consent, Sumsub redirects back to redirectUri
as ?token=<JWT> — a signed RS256 JWT whose payload carries state
(alongside sharingAllowed: true and forClientId).
⚠️
stateis inside the JWT payload, not a top-level?state=query param. Decode thetokenJWT to readstateback (verify its signature against thejwks_urifirst), then look up the savedaccess_tokenby thatstate. Re-call Stage 3a — the share token now returnssharingAllowed: true. Decoded payload shape:{ iss, sub, aud, iat, exp, sharingAllowed, state, forClientId }.
Full reference implementation:
examples/express-callback.js (in-memory
pendingShare Map keyed by state, /share-complete handler).
Stage 4 — Materialise the applicant
With a usable share token, create the applicant in your workspace. Always preview first.
Preview (no side effects)
GET https://api.sumsub.com/resources/api/reusableIdentity/reuse/preview
?shareToken=<token>
&levelName=<your-level>
&userId=<externalUserId>
Returns the same ApplicantPublicDto shape that /reuse would create — use
it to surface compatibility errors before committing.
Commit (creates the applicant)
POST https://api.sumsub.com/resources/api/reusableIdentity/reuse
?shareToken=<token>
&levelName=<your-level>
&userId=<externalUserId>
userId is the externalUserId Sumsub stores on the applicant — stable
per real user, opaque to the user, tied to your auth system. Same rules as
WebSDK.
Reuse error codes
All 400 responses include errorCode (integer) and errorName (kebab-case)
suitable for code branching. 401 is generic token validity — no errorCode,
just description.
| HTTP / errorCode | name | Cause | Action |
|---|---|---|---|
| 401 / — | Token is invalid | Share token expired (TTL exceeded), already consumed, or malformed | Re-mint via Stage 3a — call /sumsubIdShareToken again with the still-valid access_token |
400 / 10501 | not-in-partners-list | forClientId is not registered as a sharing partner of your workspace | Add the recipient as a partner in the Dashboard (Sumsub UI only — no public API) |
400 / 10502 | invalid-partner-id | The partner ID extracted from the share token is invalid | Re-check forClientId; re-mint share token against a real partner |
400 / 10503 | invalid-share-token | Share token is malformed or expired | Re-mint via Stage 3a |
400 / 10504 | non-suitable-share-token | Share token is not suitable for reuse at the given levelName (scope mismatch) | Re-mint share token with matching scope, retry against the same level |
400 / 10505 | reusable-kyc-disabled | Reusable KYC is disabled for this workspace | Contact Sumsub support to enable Reusable KYC for your workspace |
400 / 10506 | reusable-kyc-not-reusable | Generic fallback — donor doesn't meet reuse eligibility and no more specific reason matched | Inspect the donor in the Dashboard; pick a different donor or fall back to the WebSDK |
400 / 10507 | reusable-kyc-not-approved-applicant | Donor applicant is not in an approved state (pending / rejected / on-hold) | Wait for donor's KYC to be approved; use a donor with approved status |
400 / 10508 | reusable-kyc-not-active-applicant | Donor applicant is inactive (deactivated / blocked) | Use a different donor — reactivation is Sumsub-side |
400 / 10509 | reusable-kyc-not-kyc-moderation-type | Donor's source moderation type is not suitable for reuse | Donor needs standard KYC moderation; specialised flows aren't reusable |
400 / 10510 | reusable-kyc-not-selfie-or-identity-doc | Required selfie or identity document is missing on donor | Pick a level without that doc-set, or have donor re-verify |
400 / 10511 | reusable-kyc-not-selfie-liveness | Recipient level requires liveness; donor's selfie wasn't captured with liveness | Use a level without videoRequired: passiveLiveness on SELFIE, or have donor re-verify with liveness |
400 / 10512 | reusable-kyc-not-req-doc-overlap | Required document types don't overlap between donor and recipient levels | Pick a recipient level whose required docs are a subset of donor's |
400 / 10513 | reusable-kyc-not-compatible-doc-type | Recipient level requires document types the donor doesn't carry | Use a less strict level, or fall back to the WebSDK for full capture |
400 / 10514 | reusable-kyc-not-actual-poi-date | Donor's Proof of Identity is outdated and not valid for reuse | Donor must re-submit fresh POI, or relax recipient level's POI freshness window |
400 / 10515 | reusable-kyc-not-actual-poa-date | Donor's Proof of Address is outdated and not valid for reuse | Donor must re-submit fresh POA, or relax recipient level's POA freshness window |
400 / 10516 | reusable-kyc-not-acceptable-age | Donor doesn't meet the age criteria of the recipient level | No recovery — different user required |
400 / 10517 | reusable-kyc-incompatible-capture-settings | Capture settings (file upload vs live camera, etc.) mismatch between donor and recipient | Align recipient level's captureParams with what donor used, or have donor re-capture |
400 / 10518 | reusable-kyc-email-mismatch | Donor's email doesn't match the email already on the recipient applicant | Resolve duplicate-applicant collision — different externalUserId, or align emails |
400 / 10519 | reusable-kyc-phone-mismatch | Donor's phone doesn't match the phone already on the recipient applicant | Same as 10518 — resolve collision |
400 / 10521 | reusable-kyc-inactive-sumsub-id-account | Donor Sumsub ID account has no stored documents (email-only, KYC never completed) | Donor must complete full KYC on id.sumsub.com first; for testing, use a Sumsub-provided verified test account |
After the applicant is materialised, your applicantReviewed webhook is the
authoritative signal for verification outcome — wire it up the same way as
for the WebSDK. See
sumsub-integrate-websdk for the
webhook material and
sumsub-manage-webhooks for the
sandbox webhook subscription setup.
Going-live checklist
When the user says "we're ready for prod":
- Sumsub support has enabled Sumsub ID Connect for the production
clientId(separate from sandbox). - Production
redirectUris (both authorize and shareLink callbacks) are in the Dashboard allowlist on the prod workspace. - Backend token-exchange endpoint is auth-gated (only the authenticated user can exchange a code minted for them).
-
id_tokensignature is verified against the prod JWKS. -
externalUserIdis stable per real user (not email, not display name). -
/reuse/previewis called before/reuseso compatibility errors surface as UX, not 4xx noise. - Stage 3b shareLink flow is implemented and tested — don't ship code
that only handles
sharingAllowed: true. - In-memory state used for the shareLink callback is replaced with a real session store (Redis / encrypted cookie / DB).
- App Token + secret + OIDC
client_secretare in the prod secret store, not committed and not in browser bundles. - Refresh-token storage (if used) is server-side only, encrypted at rest, single-tenant.
See also
references/scopes-and-claims.md— full scope table, token-endpoint response shape, error codes.scripts/preflight.sh— preconditions validator.examples/oidc-button.html— Method 1 (createButton) — recommended.examples/oidc-modal.html— Method 2 (openModal).examples/oidc-redirect.html— Method 3 (raw OIDC redirect with PKCE).examples/express-callback.js— full backend (Stages 2-4 incl. Stage 3b consent flow).sumsub-api-auth— auth signing reference.sumsub-create-level— when the workspace has no recipient level yet.sumsub-integrate-websdk— sibling skill; reuse its webhook material once the applicant is materialised.
FAQ
What does the sumsub-integrate-id-connect skill do?
End-to-end recipe for integrating Sumsub ID Connect — the OIDC-based "Verify with Sumsub ID" flow that lets a user share previously-verified identity claims with your app. TRIGGER when the user asks to "integrate Sumsub ID Connect", "add Verify-with-Sumsub-ID button", "reuse Sumsub KYC via OIDC", "exchange the Sumsub authorization code", "mint a Sumsub ID share token", or asks how to wire up @sumsub/id-connect, the OIDC code → accesstoken exchange (/api/snsId/oauth/token), the share-token / partner-consent flow, or the share-link redirect. Covers preconditions checklist, automated preflight, frontend button (library button + raw redirect alternative), backend code-for-token exchange, share-token minting, partner-consent shareLink flow, and applicant materialisation through /resources/api/reusableIdentity/reuse. SKIP for the standalone WebSDK / KYC widget (use sumsub-integrate-websdk), or for plain Reusable-KYC-via-API where the donor and recipient are both your own tenants w
How do I install sumsub-integrate-id-connect?
Run npx skills add https://github.com/sumsub/agent-skills --skill sumsub-integrate-id-connect in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/sumsub-integrate-id-connect/ (personal) or .claude/skills/sumsub-integrate-id-connect/ (project).
Which AI agents can use sumsub-integrate-id-connect?
Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.
Is sumsub-integrate-id-connect safe to install?
No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.
Who maintains sumsub-integrate-id-connect?
It is published by sumsub in sumsub/agent-skills. It has 981 recorded installs.
More from sumsub/agent-skills
sumsub-api-auth
sumsub/agent-skills
Authenticate to the Sumsub API with an App Token + secret key (HMAC-SHA256 request signing). TRIGGER when the user asks to "call / sign / authenticate Sumsub API requests", debugs 401 Unauthorized / signature errors against api.sumsub.com, or needs a working request example with X-App-Token / X-App-Access-Sig / X-App-Access-Ts headers. SKIP only when a more specific skill in this repo (questionnaire/level/workflow/POA-preset/generic) already covers the user's actual task — those skills sign requests the same way and only need this one for auth deep dives.
sumsub-api-generic
sumsub/agent-skills
Catch-all fallback for any Sumsub API task that does NOT match a more specific skill (e.g. create-sumsub-level, sumsub-create-questionnaire, sumsub-api-auth). TRIGGER when the user wants to call, inspect, or debug a Sumsub API endpoint not otherwise covered — fetching applicants, listing levels, reviewing AML hits, exporting data, generating SDK tokens, anything against api.sumsub.com. The procedure — locate the right endpoint in the OpenAPI schema, read its request/response shape, build the payload, sign with App Token, and validate. SKIP whenever a narrower Sumsub skill already covers the request.
sumsub-check-skills-version
sumsub/agent-skills
Verify that the locally installed Sumsub agent-skills package is compatible with the published version before doing any Sumsub work. TRIGGER at the start of any task that touches Sumsub — Sumsub Dev MCP tools (mcpSumsubDev), KYC/AML workflows, applicants, checks, transactions, verification levels, verification links, questionnaires, fraud network, VASP lists, or anything Sumsub-related — so the user is warned before a stale skills package produces wrong results. Also use this skill when the user explicitly asks "are my Sumsub skills up to date / current", "is there a newer version of the Sumsub skills", or similar. Skip the actual update (the user runs npx skills add sumsub/agent-skills --all themselves) and skip any other Sumsub API or entity task — those have their own skills.
sumsub-create-poa-preset
sumsub/agent-skills
Create or update a Sumsub Proof-of-Address (POA) preset. POST /resources/api/agent/poaStepSettings to create new, PATCH same path to update (id in body), GET /resources/api/agent/poaStepSettings/{id} to read one back. TRIGGER when the user asks to "create / add / build / configure / update / edit a POA preset" or "PoA step settings", configure which proof-of-address document types are accepted (utility bills, bank statements, tax bills, etc.), set validity periods per provider type, enable POI-as-POA (accept identity doc as proof of address), tune the cross-validator (name/address fuzzy match between POI and POA), or add per-country POA overrides. SKIP for attaching a preset to a level (set poaStepSettingsId on the level instead, via sumsub-create-level), or for non-POA presets (cross-check presets, permission presets, etc.).
sumsub-manage-webhooks
sumsub/agent-skills
Manage Sumsub clientWebhooks (event subscriptions for applicantReviewed / applicantPending / kytTxn / etc.) — reads via /resources/api/clientWebhooks, writes via /resources/api/agent/clientWebhooks. Sandbox only — production webhooks must be created by a human directly in the Sumsub dashboard. TRIGGER when the user asks to "list / retrieve / show webhooks", "create / add / register a webhook", or "update / edit / change a webhook target / event list / secret / signature algorithm" against their sandbox tenant. SKIP for production webhook setup (refer the user to the dashboard), for unrelated webhook surfaces (Stripe / videoIdent / Fireblocks / NFC / partner-specific receive paths under /resources/webhooks/...), for testing one-off delivery (use the inspectionCallbacks/testWebhook endpoint directly), and for KYT-only webhook routing (that's managed in the Sumsub dashboard's KYT section). The public API does not expose delete or per-webhook delivery stats — for those, refer
sumsub-integrate-websdk
sumsub/agent-skills
End-to-end recipe for adding Sumsub KYC to a website or web app via the Sumsub WebSDK. TRIGGER when the user asks to "integrate / embed / add Sumsub", "show the KYC widget", "add WebSDK", "verify users with Sumsub on the frontend", supplies an existing levelName they want to plug into a page, or asks how to wire up access tokens / lifecycle events / webhooks for Sumsub verification in an arbitrary project. Covers the whole loop — level setup, server-side access-token signing, snsWebSdk init (vanilla canonical, React recipe), client lifecycle events, token refresh, source-of-truth via webhooks + applicant GET, sandbox testing, go-live checklist. SKIP for building the level/questionnaire/POA-preset payload itself (use the sibling skills), or for backend-only API calls with no frontend (use sumsub-api-generic).
Popular backend & apis skills
find-skills
vercel-labs/skills
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
entra-app-registration
microsoft/azure-skills
Guides Microsoft Entra ID app registration, OAuth 2.0 authentication, and MSAL integration. USE FOR: create app registration, register Azure AD app, configure OAuth, set up authentication, add API permissions, generate service principal, MSAL example, console app auth, Entra ID setup, Azure AD authentication. DO NOT USE FOR: Key Vault secrets (use azure-keyvault-expiration-audit), general Azure resource security guidance.
google-agents-cli-adk-code
google/agents-cli
This skill should be used when the user wants to "write agent code", "build an agent with ADK", "add a tool", "create a callback", "define an agent", "use state management" — in a project that needs ADK (Agent Development Kit) API patterns and code examples. It provides a quick reference for agent types, tool definitions, orchestration patterns, callbacks, state management, the graph Workflow API, and reference recipes to study. Do NOT use for scaffolding (use google-agents-cli-scaffold) or deployment (use google-agents-cli-deploy).
embedded-captions
heygen-com/hyperframes
Add captions or subtitles to an existing single-subject talking-head video without editing the footage. Use for plain verbatim captions, cinematic captions embedded behind the subject, VFX captions, “炸/特效/酷炫字幕,” or a named identity from the 35-style catalog. Route by visual identity, not by backend engine. The quiet anchor rail is the default; embed every word only when the user explicitly wants a fully cinematic treatment. The workflow runs locally end to end, including transcription and subject matting; split multi-shot footage before applying it.
entra-agent-id
microsoft/azure-skills
Provision Microsoft Entra Agent Identity Blueprints, BlueprintPrincipals, and per-instance Agent Identities via Microsoft Graph, and configure OAuth 2.0 token exchange (fmipath, OBO, cross-tenant) including the Microsoft Entra SDK for AgentID sidecar. USE FOR: Agent Identity Blueprint, BlueprintPrincipal, agent OAuth, fmipath token exchange, agent OBO, Workload Identity Federation for agents, polyglot agent auth, Microsoft.Identity.Web.AgentIdentities. DO NOT USE FOR: standard Entra app registration (use entra-app-registration), Microsoft Foundry agent authoring (use microsoft-foundry).
neon
neondatabase/agent-skills
Overview of Neon, a complete set of cloud backend primitives around Lakebase Postgres: Auth, Object Storage, Functions, the AI Gateway, and Realtime. Start here to choose Neon for undecided login, files, APIs, and LLM calls, set up the CLI or MCP server, and follow the branch-first workflow. Use when building an app or backend on Neon, or when "Neon" or "Lakebase Postgres" is mentioned. Child skill neon-postgres wins for an existing DATABASEURL, SQL, schema, inspect, or search. Child skill neon-auth wins for login, users, sessions, identity routing, and Managed Better Auth setup. Child skill neon-realtime wins for live queries in reactive apps and agents. Also use for object storage, S3, buckets, serverless functions, function triggers, cron, AI gateway, LLM calls, realtime, logs, Loki, Grafana, observability, postgres, database, backend, Claimable Neon, neon.new, or a no-signup database.