GHSA-HP6V-6JW7-GV2F
Vulnerability from github – Published: 2026-07-24 21:17 – Updated: 2026-07-24 21:17Summary
Budibase's OIDC SSO login links an incoming SSO identity to an existing Budibase account by email address alone, without ever checking the email_verified claim of the OIDC ID token. Budibase first tries to match the IdP sub; when that misses (any fresh attacker IdP account) it silently falls back to matching by the email claim and merges into the existing account by email, preserving that account's _id and roles. Because the email_verified flag is never read, an attacker who can make a configured/trusted IdP emit a token carrying email = <victim> with email_verified = false is logged into Budibase as the victim, inheriting the victim's roles (including global admin/builder). Per OIDC Core §5.7 the email claim MUST NOT be used as an identity key unless email_verified is true; Budibase effectively delegates all account-linking trust to every configured IdP's email-verification policy while checking nothing itself. Full account takeover of any existing Budibase user, including the instance owner.
Details
The OIDC verify callback extracts the email and never consults email_verified:
- packages/backend-core/src/middleware/passport/sso/oidc.ts:59 — email: getEmail(profile, jwtClaims).
- getEmail (oidc.ts:113-135) returns profile._json.email ->jwtClaims.email -> preferred_username. No email_verified check.
- buildJwtClaims (oidc.ts:99-107) assembles claims from _json.email/emails[0].value — no verification flag is read. grep -r email_verified packages/ -> 0 hits.
The email is then used as the account-linking key:
- sso.authenticate(...) -> packages/backend-core/src/middleware/passport/sso/sso.ts:
- :38,44 users.getById(generateGlobalUserID(details.userId)) keyed on the IdP sub; for a fresh attacker IdP account this 404s and is swallowed (:45-54).
- :57-59 fallback: dbUser = await users.getGlobalUserByEmail(details.email) -> loads the victim's account (victim _id + roles) purely by email (packages/backend-core/src/users/users.ts:100-124, USER_BY_EMAIL view, no binding to the IdP sub).
- syncUser(...) (sso.ts:80,102-138) spreads ...user, preserving the victim _id/tenantId/roles; only overwrites provider fields.
- UserDB.save (packages/backend-core/src/users/db.ts:235): because ssoUser._id is the victim's, the _id branch runs (:253), getById(_id) matches the victim (:256), the "Email address cannot be changed" guard (:257-259) does not fire (dbUser.email === email), and the EmailUnavailableError guard (:269-275) is skipped (it only runs in the !dbUser branch). The merge proceeds silently; a session JWT is issued for the victim.
Per OIDC Core §5.7, the email claim MUST NOT be used as an identity key unless email_verified is true. Budibase never reads the flag.
Preconditions (attack requirement — AT:P): the attacker must be able to authenticate through an IdP that the Budibase instance trusts AND get that IdP to assert the victim's email with email_verified = false. This is reachable, not exotic:
- Self-registration with an unverified email — Keycloak and Authentik ship with "Verify Email" OFF by default; if the trusted IdP allows public sign-up, the attacker registers a new account and simply enters email = <victim> at sign-up. No confirmation email is needed — the IdP stores and asserts it unverified.
- Self-service profile editing — many IdPs let a logged-in user change their own email without forced re-verification.
- Attacker-operated / federated IdP or permissive social login — where the attacker controls or influences a trusted provider, or the provider asserts a user-typed (unverified) email.
It is not exploitable through a strict corporate IdP that enforces email verification (there email_verified = true and the attacker cannot claim the victim's address) — which is exactly why Budibase must check the flag rather than assume every configured IdP enforces it. The defect is unconditional on the Budibase side; the attack requirement is purely the (default, common) IdP email policy.
PoC
Reproduced live on Budibase 3.39.14 (self-hosted, community license) against a stock Keycloak 26 realm budi with default "Verify Email" = off; OIDC client registered and activated in Budibase.
Setup: a pre-existing victim global-admin Budibase account victim@stand.local (_id = us_1ab2dfcf…, local account, no IdP link). The attacker owns their own IdP account (attacker, distinct sub) and can set its email attribute unverified.
Step 1 — the IdP asserts the claim (proves email_verified=false):
POST /realms/budi/protocol/openid-connect/token (Keycloak)
grant_type=password&client_id=budibase&client_secret=…&username=attacker&password=Attacker123!&scope=openid email profile
-> id_token payload: { "sub":"3cf58c45-…", "preferred_username":"attacker",
"email":"victim@stand.local", "email_verified":false }
The authenticated principal is provably attacker (its own sub/preferred_username/password), merely claiming the victim's email, unverified.
Step 2 — drive the standard OIDC flow as attacker:
GET /api/global/auth/default/oidc/configs/kc-oidc-1 -> IdP login as attacker/Attacker123! -> GET /api/global/auth/oidc/callback?code=…&state=….
Step 3 — result (takeover): Budibase sets budibase:auth to a session JWT
{ "userId":"us_1ab2dfcf…", "email":"victim@stand.local", "tenantId":"default" }, and
GET /api/global/self returns the victim: _id = us_1ab2dfcf…, admin.global = true, builder.global = true, providerType = oidc. The attacker authenticated as a different IdP principal with an unverified email yet now holds a full global-admin session for the victim.
Negative control (proves the email claim is the cause, not a normal self-login):
A second attacker attacker2 with a benign unverified email attacker2@evil.local (matching no Budibase user) runs the identical flow:
id_token: { "sub":"55794bf2-…", "preferred_username":"attacker2", "email":"attacker2@evil.local", "email_verified":false }
-> budibase:auth: { "userId":"us_55794bf2-…" } (a NEW account, _id derived from the IdP sub)
-> /api/global/self: { "_id":"us_55794bf2-…", "email":"attacker2@evil.local", admin.global: null, builder.global: null }
With a benign email the attacker gets their own new low-privilege account; only when the email claim equals the victim's does the same flow yield the victim's admin account. Same self-authentication, single variable changed = the unverified-email merge is the vulnerability.
Impact
Takeover of any existing Budibase account by email, including the instance owner / global admin -> full control of the tenant (apps, datasources, automations, user management, stored datasource credentials). The attacker authenticates as their own (different) IdP principal and ends up holding the victim's session and roles. The only requirement beyond a trusted IdP login is that the IdP assert the victim's email unverified — the default for a freshly-created Keycloak/Authentik realm and common in social logins (see Preconditions). Any deployment that trusts an OIDC IdP without enforced email verification is exposed; the Budibase-side flaw (ignoring email_verified) is unconditional.
Remediation
Primary fix: in the OIDC verify path, require email_verified === true before using email to look up / link an existing account; otherwise reject the login (or fall back to sub-only matching and never merge into a pre-existing local/SSO account). Concretely, thread the email_verified claim through buildJwtClaims/getEmail (oidc.ts) and gate the getGlobalUserByEmail fallback in sso.ts:57-59 on it.
Audit the whole class: apply the same email_verified (and, for SAML, EmailVerified/assertion-signature) gate to every SSO strategy that links by email — OIDC, SAML, and any social provider — not only the Google strategy (which already passes requireLocalAccount=true). Email-based account linking anywhere must require a verified email.
Defense-in-depth for operators who cannot patch immediately:
- On the IdP, enable "Verify Email" / require verified email before issuing tokens (Keycloak: realm -> Login -> Verify Email = on), and restrict which email domains the IdP will assert.
- Prefer sub-based account mapping over email in the IdP/Budibase mapping config where available.
- Audit existing accounts for unexpected OIDC links to privileged users; rotate sessions.
{
"affected": [
{
"package": {
"ecosystem": "npm",
"name": "@budibase/server"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "3.38.1"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [],
"database_specific": {
"cwe_ids": [
"CWE-287"
],
"github_reviewed": true,
"github_reviewed_at": "2026-07-24T21:17:39Z",
"nvd_published_at": null,
"severity": "CRITICAL"
},
"details": "### Summary\nBudibase\u0027s OIDC SSO login links an incoming SSO identity to an existing Budibase account **by email address alone**, without ever checking the `email_verified` claim of the OIDC ID token. Budibase first tries to match the IdP `sub`; when that misses (any fresh attacker IdP account) it silently falls back to matching by the `email` claim and **merges into the existing account by email**, preserving that account\u0027s `_id` and roles. Because the `email_verified` flag is never read, an attacker who can make a **configured/trusted** IdP emit a token carrying `email = \u003cvictim\u003e` with `email_verified = false` is logged into Budibase **as the victim**, inheriting the victim\u0027s roles (including global admin/builder). Per OIDC Core \u00a75.7 the `email` claim MUST NOT be used as an identity key unless `email_verified` is `true`; Budibase effectively delegates all account-linking trust to every configured IdP\u0027s email-verification policy while checking nothing itself. Full account takeover of any existing Budibase user, including the instance owner.\n\n### Details\nThe OIDC verify callback extracts the email and never consults `email_verified`:\n- `packages/backend-core/src/middleware/passport/sso/oidc.ts:59` \u2014 `email: getEmail(profile, jwtClaims)`.\n- `getEmail` (`oidc.ts:113-135`) returns `profile._json.email` -\u003e`jwtClaims.email` -\u003e `preferred_username`. **No `email_verified` check.**\n- `buildJwtClaims` (`oidc.ts:99-107`) assembles claims from `_json.email`/`emails[0].value` \u2014 no verification flag is read. `grep -r email_verified packages/` -\u003e 0 hits.\n\nThe email is then used as the **account-linking key**:\n- `sso.authenticate(...)` -\u003e `packages/backend-core/src/middleware/passport/sso/sso.ts`:\n - `:38,44` `users.getById(generateGlobalUserID(details.userId))` keyed on the IdP `sub`; for a fresh attacker IdP account this 404s and is swallowed (`:45-54`).\n - `:57-59` **fallback:** `dbUser = await users.getGlobalUserByEmail(details.email)` -\u003e loads the **victim\u0027s** account (victim `_id` + roles) purely by email (`packages/backend-core/src/users/users.ts:100-124`, `USER_BY_EMAIL` view, no binding to the IdP `sub`).\n - `syncUser(...)` (`sso.ts:80,102-138`) spreads `...user`, preserving the victim `_id`/`tenantId`/`roles`; only overwrites provider fields.\n- `UserDB.save` (`packages/backend-core/src/users/db.ts:235`): because `ssoUser._id` is the victim\u0027s, the `_id` branch runs (`:253`), `getById(_id)` matches the victim (`:256`), the \"Email address cannot be changed\" guard (`:257-259`) does not fire (`dbUser.email === email`), and the `EmailUnavailableError` guard (`:269-275`) is skipped (it only runs in the `!dbUser` branch). The merge proceeds silently; a session JWT is issued for the victim.\n\nPer OIDC Core \u00a75.7, the `email` claim MUST NOT be used as an identity key unless `email_verified` is `true`. Budibase never reads the flag.\n\n**Preconditions (attack requirement \u2014 `AT:P`):** the attacker must be able to authenticate through an IdP that the Budibase instance **trusts** AND get that IdP to assert the victim\u0027s email with `email_verified = false`. This is reachable, not exotic:\n- **Self-registration with an unverified email \u2014 Keycloak and Authentik ship with *\"Verify Email\" OFF by default*; if the trusted IdP allows public sign-up, the attacker registers a new account and simply enters `email = \u003cvictim\u003e` at sign-up. No confirmation email is needed \u2014 the IdP stores and asserts it unverified.**\n- **Self-service profile editing** \u2014 many IdPs let a logged-in user change their own email without forced re-verification.\n- **Attacker-operated / federated IdP or permissive social login** \u2014 where the attacker controls or influences a trusted provider, or the provider asserts a user-typed (unverified) email.\nIt is **not** exploitable through a strict corporate IdP that enforces email verification (there `email_verified = true` and the attacker cannot claim the victim\u0027s address) \u2014 which is exactly why Budibase must check the flag rather than assume every configured IdP enforces it. The defect is unconditional on the Budibase side; the attack requirement is purely the (default, common) IdP email policy.\n\n### PoC\nReproduced live on Budibase `3.39.14` (self-hosted, community license) against a stock **Keycloak 26** realm `budi` with default \"Verify Email\" = off; OIDC client registered and activated in Budibase.\n\n**Setup:** a pre-existing **victim** global-admin Budibase account `victim@stand.local` (`_id = us_1ab2dfcf\u2026`, local account, *no IdP link*). The attacker owns their **own** IdP account (`attacker`, distinct `sub`) and can set its email attribute unverified.\n\n**Step 1 \u2014 the IdP asserts the claim (proves `email_verified=false`):**\n```\nPOST /realms/budi/protocol/openid-connect/token (Keycloak)\ngrant_type=password\u0026client_id=budibase\u0026client_secret=\u2026\u0026username=attacker\u0026password=Attacker123!\u0026scope=openid email profile\n-\u003e id_token payload: { \"sub\":\"3cf58c45-\u2026\", \"preferred_username\":\"attacker\",\n \"email\":\"victim@stand.local\", \"email_verified\":false }\n```\nThe authenticated principal is provably **`attacker`** (its own `sub`/`preferred_username`/password), merely *claiming* the victim\u0027s email, unverified.\n\n\u003cimg width=\"1484\" height=\"685\" alt=\"image\" src=\"https://github.com/user-attachments/assets/cdddac3c-b7c0-4c59-aaec-9092706a7ad8\" /\u003e\n\n\n**Step 2 \u2014 drive the standard OIDC flow** as `attacker`:\n`GET /api/global/auth/default/oidc/configs/kc-oidc-1` -\u003e IdP login as `attacker`/`Attacker123!` -\u003e `GET /api/global/auth/oidc/callback?code=\u2026\u0026state=\u2026`.\n\n\u003cimg width=\"1145\" height=\"454\" alt=\"image\" src=\"https://github.com/user-attachments/assets/ac0a73d0-c5fc-4d0f-b9b0-95f6273b99f7\" /\u003e\n\n\u003cimg width=\"1172\" height=\"445\" alt=\"image\" src=\"https://github.com/user-attachments/assets/a4b27a7e-4293-4caf-957f-d27c54ea461a\" /\u003e\n\n\u003cimg width=\"1157\" height=\"648\" alt=\"image\" src=\"https://github.com/user-attachments/assets/06699774-92b0-4163-a171-9a30bc877ecc\" /\u003e\n\n\u003cimg width=\"1201\" height=\"525\" alt=\"image\" src=\"https://github.com/user-attachments/assets/92a92fcf-9d29-42ca-921d-de6fbd78198f\" /\u003e\n\n\n\n**Step 3 \u2014 result (takeover):** Budibase sets `budibase:auth` to a session JWT\n`{ \"userId\":\"us_1ab2dfcf\u2026\", \"email\":\"victim@stand.local\", \"tenantId\":\"default\" }`, and\n`GET /api/global/self` returns the **victim**: `_id = us_1ab2dfcf\u2026`, `admin.global = true`, `builder.global = true`, `providerType = oidc`. The attacker authenticated as a *different* IdP principal with an *unverified* email yet now holds a full global-admin session for the victim.\n\n\u003cimg width=\"1001\" height=\"456\" alt=\"image\" src=\"https://github.com/user-attachments/assets/c8dfc25f-aed7-4cd7-9323-f029e4d1a725\" /\u003e\n\n\n**Negative control (proves the email claim is the cause, not a normal self-login):**\nA second attacker `attacker2` with a **benign** unverified email `attacker2@evil.local` (matching no Budibase user) runs the *identical* flow:\n```\nid_token: { \"sub\":\"55794bf2-\u2026\", \"preferred_username\":\"attacker2\", \"email\":\"attacker2@evil.local\", \"email_verified\":false }\n-\u003e budibase:auth: { \"userId\":\"us_55794bf2-\u2026\" } (a NEW account, _id derived from the IdP sub)\n-\u003e /api/global/self: { \"_id\":\"us_55794bf2-\u2026\", \"email\":\"attacker2@evil.local\", admin.global: null, builder.global: null }\n```\n\u003cimg width=\"1153\" height=\"489\" alt=\"image\" src=\"https://github.com/user-attachments/assets/8f6cb849-e6f3-49f1-b576-8aeba026a3be\" /\u003e\n\n\u003cimg width=\"1089\" height=\"466\" alt=\"image\" src=\"https://github.com/user-attachments/assets/a892ec3e-3753-4bc0-9eee-9f0613b2d92e\" /\u003e\n\n\u003cimg width=\"826\" height=\"532\" alt=\"image\" src=\"https://github.com/user-attachments/assets/cc30bd9c-1bdb-4bea-934f-a8b3e228940e\" /\u003e\n\n\nWith a benign email the attacker gets **their own new low-privilege account**; only when the email claim equals the victim\u0027s does the same flow yield the **victim\u0027s admin account**. Same self-authentication, single variable changed = the unverified-email merge is the vulnerability.\n\n### Impact\nTakeover of any existing Budibase account by email, including the instance owner / global admin -\u003e full control of the tenant (apps, datasources, automations, user management, stored datasource credentials). The attacker authenticates as their *own* (different) IdP principal and ends up holding the victim\u0027s session and roles. The only requirement beyond a trusted IdP login is that the IdP assert the victim\u0027s email unverified \u2014 the default for a freshly-created Keycloak/Authentik realm and common in social logins (see Preconditions). Any deployment that trusts an OIDC IdP without enforced email verification is exposed; the Budibase-side flaw (ignoring `email_verified`) is unconditional.\n\n### Remediation\n**Primary fix:** in the OIDC verify path, require `email_verified === true` before using `email` to look up / link an existing account; otherwise reject the login (or fall back to `sub`-only matching and never merge into a pre-existing local/SSO account). Concretely, thread the `email_verified` claim through `buildJwtClaims`/`getEmail` (`oidc.ts`) and gate the `getGlobalUserByEmail` fallback in `sso.ts:57-59` on it.\n\n**Audit the whole class:** apply the same `email_verified` (and, for SAML, `EmailVerified`/assertion-signature) gate to every SSO strategy that links by email \u2014 OIDC, SAML, and any social provider \u2014 not only the Google strategy (which already passes `requireLocalAccount=true`). Email-based account linking anywhere must require a verified email.\n\n**Defense-in-depth for operators who cannot patch immediately:**\n- On the IdP, enable \"Verify Email\" / require verified email before issuing tokens (Keycloak: realm -\u003e Login -\u003e Verify Email = on), and restrict which email domains the IdP will assert.\n- Prefer `sub`-based account mapping over email in the IdP/Budibase mapping config where available.\n- Audit existing accounts for unexpected OIDC links to privileged users; rotate sessions.",
"id": "GHSA-hp6v-6jw7-gv2f",
"modified": "2026-07-24T21:17:39Z",
"published": "2026-07-24T21:17:39Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/Budibase/budibase/security/advisories/GHSA-hp6v-6jw7-gv2f"
},
{
"type": "WEB",
"url": "https://github.com/Budibase/budibase/commit/9ecd0048d9c3ae0ee9bd0e6204c621794dd1a4d3"
},
{
"type": "PACKAGE",
"url": "https://github.com/Budibase/budibase"
},
{
"type": "WEB",
"url": "https://github.com/Budibase/budibase/releases/tag/3.39.30"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:4.0/AV:N/AC:L/AT:P/PR:L/UI:N/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H",
"type": "CVSS_V4"
}
],
"summary": " Budibase: OIDC SSO account takeover: incoming identity linked by email without checking email_verified"
}
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.