* fix(security): bind api_keys to the caller, lock hash-as-bearer RPCs and provider token tables Security audit 2026-09-01, critical items. - api_keys INSERT requires user_id = auth.uid() again (an admin could forge a key for any co-member and act as them in every company they belong to); SELECT is own-keys-or-admin; a BEFORE trigger freezes the identity and credential columns against user-session UPDATEs. - rotate_mcp_refresh_token and validate_and_increment_api_key become service_role only: they match rows by a presented SHA-256, so a hash readable by co-members was a bearer credential. - validate_and_increment_api_key fails closed when the key's user is no longer a member of the key's company. - provider_consent_tokens and provider_otc: the DELETE policies collapsed to "caller has any team row" (correlated subquery on a non-existent team_members.company_id). All member policies dropped; service_role only, matching every existing code path. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): role gates, ownership guards and posting integrity in the database Security audit 2026-09-01, high items at the database layer. - One table-level guard, enforce_company_writer_role(), blocks the read-only viewer role on 55 company-scoped tables including through the 15 membership-only SECURITY DEFINER writers. Keyed on the JWT role claim so it fires inside definer bodies; no-op for service_role and trigger cascades. - company_members user_id/company_id immutable from user sessions; invitations can never grant owner; team_members gains a transition guard (admins keep non-owner role moves); companies team_id and archiving are owner-only and team attachment needs team membership. - Direct statements (current_user = authenticated) can no longer insert posted headers, add lines under posted verifikat, or post a draft with a voucher number the sequence never issued. Sanctioned RPCs run as the definer and are untouched; the engine's own draft-then-post shapes still pass. - create_document_version refuses viewers and foreign storage paths; validate_version_chain needs membership and loses anon EXECUTE; match_documents / match_booking_templates lose anon; cron maintenance RPCs become service_role only; the production-only seed_asset_categories is dropped. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * build: pin tsx as an exact devDependency instead of fetching it with npx at build time prebuild ran "npx tsx" with no lockfile entry, so every Vercel, Docker and CI build downloaded tsx@latest and its transitive tree from the registry with no integrity check, inside the build environment. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): refuse the viewer role on API-key and MCP write paths The v1 wrapper and the MCP company routing checked company membership but never role, and both run as service role, so a read-only viewer holding an API key could post vouchers and change settings through the API. Mutating methods and non-read scopes now return 403 ROLE_READ_ONLY for viewers on v1; MCP write tools refuse viewers the same way. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): stop serving uploaded SVG, XML and HTML as executable content on the app origin Uploads persisted the browser-declared mime type and the inline proxy served it verbatim, sandboxing only text/html; the storage proxy forwarded the uploader's Content-Type. Any writer, or any Peppol sender, could plant a scripted SVG or XHTML that executed on app.gnubok.se. - inline route: allow-list of natively safe types (PDF, raster images) served as before; everything else gets the opaque sandbox CSP. - storage proxy: octet-stream + attachment + sandbox unless the DB mime for the key is on the allow-list. - document-service: the stored mime is the magic-byte validated type. - logo upload: magic-byte validation, SVG refused. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): byrå brand logo upload decides the type by magic bytes and drops SVG Same pattern as the company logo route: the logos bucket is public, so a scripted SVG (or anything declared as an image) must never land there. The upload pickers stop advertising SVG. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): bind Enable Banking, Stripe and WooCommerce callbacks to the initiating user The callbacks resolved the pending row by oauth_state alone, so a victim who completed an attacker-initiated consent had their bank account, merchant account or store attached to the attacker's company. requireFlowInitiator() now requires the cookie session of the user who started the flow: no session redirects to login with the callback URL preserved, a different user is refused and nothing is exchanged. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): guard tenant-controlled outbound fetches and surface the disabled rate limiter WooCommerce and Shopify syncs fetched a member-editable store URL with plain fetch() and redirect following under the service role, and the invoice PDF renderer fetched company_settings.logo_url unguarded. All three go through a new safeFetch() (public-IP validation via url-guard, https only, redirect: 'manual', body size cap) and re-normalise the stored host at use time. checkRateLimit() keeps failing open on hosted but logs one error per process when Upstash is not configured and exports isRateLimiterConfigured(). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): decide the API MFA gate from server-authenticated factors, not the session cookie getAuthenticatorAssuranceLevel() without arguments derives nextLevel from session.user.factors, which comes from the unsigned sb-*-auth-token cookie. Deleting factors from the cookie made an enrolled account look like it had nothing to step up to, on every /api route and in requireAuth. Both gates now read factors from the getUser() result or listFactors() and the level from the verified JWT claim, and fail closed on errors. Page-branch gate hardened the same way. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): bind Fortnox/Visma, Gmail and Skatteverket callbacks to the initiating user The arcim-migration callback exchanged the provider code onto whatever consent the one-time state named, with no check of who completed the flow and no org-number comparison, so a phished Fortnox admin handed their ledger to the attacker's company. provider_otc now records the initiating user (migration 20260902100000); the callback requires that session and, after the exchange, refuses a provider company whose org number differs from the consent's company. The Gmail and Skatteverket callbacks enforce the same initiator check. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): BankID signup confirms the email before linking the identity Signup created an email-confirmed, MFA-exempt account for any address the caller typed and returned a magic link, so an attacker could pre-register a victim's email and keep a permanent BankID login into the account the victim later adopted. The user is now created unconfirmed, the identity carries email_verified_at NULL (migration 20260902101000), bankid_linked is not set until the mailed confirmation is clicked, and BankID login of a pending identity is refused with the confirmation re-sent. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): bind MCP OAuth redirect URIs to the consenting user and cap scopes A user-registered redirect URI was allowlisted globally, the consent page named no client, and all scopes were pre-checked, so one phishing link handed an attacker a full-scope key for the victim's company. Registered URIs now resolve only for the registrant or a colleague sharing a company; the consent page shows the client identity and redirect host; non-built-in clients default to read-only pre-checks; scopes are capped by the user's role (viewer: read only) at consent and at /token. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(auth): client follow-ups for BankID confirmation, callback mismatch copy and decision log - register client handles the new confirmation_sent response from BankID signup with the existing inbox screen instead of calling verifyOtp. - BankID login surfaces the email_unconfirmed explanation. - WooCommerce settings map woocommerce_error=wrong_user to its own copy. - Logo help text no longer advertises SVG. - DECISIONS.md records the audit remediation choices. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(mcp-oauth): literal SoD columns in the api_keys insert so the phantom-column scanner resolves them Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * test(logo): type the upload fixtures as Uint8Array<ArrayBuffer> so they are valid BlobParts Fixes the typecheck ratchet on PR #2155 and ratchets the baseline down by the one legacy error the change removed. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
375 lines
13 KiB
TypeScript
375 lines
13 KiB
TypeScript
import { isUnsafeUrlError, safeFetch } from '@/lib/http/safe-fetch'
|
|
import type { WooOrder, WooRefund, WooStoreInfo } from '../types'
|
|
|
|
/**
|
|
* Minimal WooCommerce REST API (wc/v3) client for the order feed.
|
|
*
|
|
* Auth is HTTP Basic (consumer key as username, secret as password) over
|
|
* HTTPS only. Some hosts (Apache CGI, security plugins) strip the
|
|
* Authorization header, so a 401 is retried once with the documented
|
|
* query-string credential fallback; that fallback is why plain-http stores
|
|
* are refused outright (keys in a cleartext URL are a credentials leak).
|
|
*
|
|
* The store URL is tenant input that the server connects to, and members can
|
|
* write `woocommerce_connections.store_url` directly through PostgREST
|
|
* (bypassing the connect route's normalisation), so every request here
|
|
* re-normalises the stored URL and goes through `safeFetch`: public
|
|
* addresses only, checked at request time, and no redirects followed. The
|
|
* nightly cron runs this under the service role, which is exactly the
|
|
* network position an SSRF would want.
|
|
*
|
|
* Typical WooCommerce hosts are slow shared PHP boxes: requests run
|
|
* sequentially, pages are capped at 100 rows, and 429/5xx responses get a
|
|
* short exponential backoff before the error is surfaced.
|
|
*/
|
|
|
|
const REQUEST_TIMEOUT_MS = 30_000
|
|
const RETRYABLE_STATUS = new Set([429, 502, 503, 504])
|
|
const RETRY_DELAYS_MS = [1_000, 3_000]
|
|
/** wc/v3 hard maximum for per_page. */
|
|
export const WC_PAGE_SIZE = 100
|
|
|
|
export interface WooCredentials {
|
|
storeUrl: string
|
|
consumerKey: string
|
|
consumerSecret: string
|
|
}
|
|
|
|
export class WooCommerceApiError extends Error {
|
|
constructor(
|
|
message: string,
|
|
/** HTTP status, or 0 for network-level failures. */
|
|
readonly status: number,
|
|
/** WooCommerce error code (e.g. woocommerce_rest_cannot_view), if any. */
|
|
readonly wooCode: string | null = null,
|
|
) {
|
|
super(message)
|
|
this.name = 'WooCommerceApiError'
|
|
}
|
|
}
|
|
|
|
/**
|
|
* Whether an API error means the credentials themselves are dead (key deleted
|
|
* or demoted in wp-admin), as opposed to a transient failure. Used to flip a
|
|
* connection to status 'revoked' so the UI offers a reconnect instead of the
|
|
* cron retrying forever.
|
|
*/
|
|
export function isRevokedCredentialsError(error: unknown): boolean {
|
|
if (!(error instanceof WooCommerceApiError)) return false
|
|
return error.status === 401 || error.status === 403
|
|
}
|
|
|
|
/**
|
|
* Hostnames the server must never fetch: the store URL is user input that we
|
|
* probe server-side, so loopback/link-local/private ranges and internal
|
|
* naming conventions are refused outright (SSRF guard). Hostname-level only,
|
|
* and cheap enough to run synchronously at connect time; a public DNS name
|
|
* resolving to a private address is caught later by `safeFetch`, which
|
|
* resolves and classifies every A/AAAA record at request time.
|
|
*/
|
|
function isDisallowedHost(hostname: string): boolean {
|
|
const h = hostname.toLowerCase()
|
|
if (h === 'localhost' || h.endsWith('.localhost')) return true
|
|
if (h.endsWith('.local') || h.endsWith('.internal')) return true
|
|
// IPv6 literals (URL.hostname strips the brackets): never a real store.
|
|
if (h.includes(':')) return true
|
|
const v4 = /^(\d{1,3})\.(\d{1,3})\.(\d{1,3})\.(\d{1,3})$/.exec(h)
|
|
if (v4) {
|
|
const a = Number(v4[1])
|
|
const b = Number(v4[2])
|
|
if (a === 0 || a === 10 || a === 127) return true
|
|
if (a === 169 && b === 254) return true
|
|
if (a === 172 && b >= 16 && b <= 31) return true
|
|
if (a === 192 && b === 168) return true
|
|
}
|
|
return false
|
|
}
|
|
|
|
/**
|
|
* Normalize and validate a user-entered store URL to an https origin plus
|
|
* optional subdirectory path (WordPress installs under a path are common),
|
|
* lowercased host, no trailing slash, no query/fragment/credentials, and no
|
|
* private/internal hosts. Returns null for anything invalid, including
|
|
* plain http.
|
|
*/
|
|
export function normalizeStoreUrl(input: string): string | null {
|
|
const trimmed = input.trim()
|
|
if (!trimmed) return null
|
|
const withScheme = /^[a-z][a-z0-9+.-]*:\/\//i.test(trimmed) ? trimmed : `https://${trimmed}`
|
|
let url: URL
|
|
try {
|
|
url = new URL(withScheme)
|
|
} catch {
|
|
return null
|
|
}
|
|
if (url.protocol !== 'https:') return null
|
|
if (url.username || url.password || url.search || url.hash) return null
|
|
if (isDisallowedHost(url.hostname)) return null
|
|
const path = url.pathname.replace(/\/+$/, '')
|
|
return `https://${url.host.toLowerCase()}${path}`
|
|
}
|
|
|
|
/** Error code on a WooCommerceApiError when the stored store URL fails re-normalisation. */
|
|
export const INVALID_STORE_URL_CODE = 'accounted_invalid_store_url'
|
|
/** Error code on a WooCommerceApiError when the SSRF guard refused to connect. */
|
|
export const UNSAFE_STORE_URL_CODE = 'accounted_unsafe_store_url'
|
|
|
|
/**
|
|
* Re-run the connect-time normalisation on the STORED store URL at use time.
|
|
* The connect route normalises what the user typed, but a member can PATCH
|
|
* `store_url` straight into the row through PostgREST, so the value in the
|
|
* database is not trusted to still be an https public-host origin. A row that
|
|
* fails here is refused with a clear, non-retryable error instead of fetched.
|
|
*/
|
|
function storeOriginOf(creds: WooCredentials): string {
|
|
const normalized = normalizeStoreUrl(creds.storeUrl)
|
|
if (!normalized) {
|
|
throw new WooCommerceApiError(
|
|
`WooCommerce store URL is not a valid https store address (${creds.storeUrl}); reconnect the store`,
|
|
0,
|
|
INVALID_STORE_URL_CODE,
|
|
)
|
|
}
|
|
return normalized
|
|
}
|
|
|
|
function sleep(ms: number): Promise<void> {
|
|
return new Promise((resolve) => setTimeout(resolve, ms))
|
|
}
|
|
|
|
function buildUrl(
|
|
creds: WooCredentials,
|
|
path: string,
|
|
params: Record<string, string>,
|
|
credentialsInQuery: boolean,
|
|
): string {
|
|
const url = new URL(`${storeOriginOf(creds)}/wp-json/wc/v3${path}`)
|
|
for (const [key, value] of Object.entries(params)) url.searchParams.set(key, value)
|
|
if (credentialsInQuery) {
|
|
url.searchParams.set('consumer_key', creds.consumerKey)
|
|
url.searchParams.set('consumer_secret', creds.consumerSecret)
|
|
}
|
|
return url.toString()
|
|
}
|
|
|
|
async function requestOnce(
|
|
creds: WooCredentials,
|
|
path: string,
|
|
params: Record<string, string>,
|
|
credentialsInQuery: boolean,
|
|
): Promise<Response> {
|
|
const headers: Record<string, string> = { Accept: 'application/json' }
|
|
if (!credentialsInQuery) {
|
|
const basic = Buffer.from(`${creds.consumerKey}:${creds.consumerSecret}`).toString('base64')
|
|
headers.Authorization = `Basic ${basic}`
|
|
}
|
|
// safeFetch: public address only (checked now, not at connect time), no
|
|
// redirects. A 3xx from the store is a failure, never a hop.
|
|
return safeFetch(buildUrl(creds, path, params, credentialsInQuery), {
|
|
headers,
|
|
signal: AbortSignal.timeout(REQUEST_TIMEOUT_MS),
|
|
})
|
|
}
|
|
|
|
/**
|
|
* Map a failure from requestOnce to the error the retry loop should see.
|
|
* Guard refusals (bad stored URL, private address, redirect) are terminal:
|
|
* retrying the same URL cannot succeed and must not spend the backoff budget.
|
|
*/
|
|
function asTerminalGuardError(err: unknown): WooCommerceApiError | null {
|
|
if (err instanceof WooCommerceApiError) return err
|
|
if (isUnsafeUrlError(err)) {
|
|
return new WooCommerceApiError(
|
|
`WooCommerce store refused by outbound URL guard: ${err.detail}`,
|
|
0,
|
|
UNSAFE_STORE_URL_CODE,
|
|
)
|
|
}
|
|
return null
|
|
}
|
|
|
|
async function parseError(response: Response): Promise<WooCommerceApiError> {
|
|
let wooCode: string | null = null
|
|
let detail = ''
|
|
try {
|
|
const body = (await response.json()) as { code?: string; message?: string }
|
|
wooCode = body.code ?? null
|
|
detail = body.message ?? ''
|
|
} catch {
|
|
// Non-JSON error body (host error page); the status is enough.
|
|
}
|
|
return new WooCommerceApiError(
|
|
`WooCommerce API ${response.status}${detail ? `: ${detail}` : ''}`,
|
|
response.status,
|
|
wooCode,
|
|
)
|
|
}
|
|
|
|
/**
|
|
* GET a wc/v3 path. Retries the header-stripped-auth case (401 → query-string
|
|
* credentials) once, and 429/5xx with a short backoff.
|
|
*/
|
|
export async function wcGet<T>(
|
|
creds: WooCredentials,
|
|
path: string,
|
|
params: Record<string, string> = {},
|
|
): Promise<T> {
|
|
let credentialsInQuery = false
|
|
let lastError: unknown
|
|
for (let attempt = 0; attempt <= RETRY_DELAYS_MS.length; attempt++) {
|
|
let response: Response
|
|
try {
|
|
response = await requestOnce(creds, path, params, credentialsInQuery)
|
|
} catch (err) {
|
|
const terminal = asTerminalGuardError(err)
|
|
if (terminal) throw terminal
|
|
// Network/timeout errors: retry on the same backoff schedule.
|
|
lastError = new WooCommerceApiError(
|
|
`WooCommerce request failed: ${err instanceof Error ? err.message : String(err)}`,
|
|
0,
|
|
)
|
|
if (attempt < RETRY_DELAYS_MS.length) {
|
|
await sleep(RETRY_DELAYS_MS[attempt])
|
|
continue
|
|
}
|
|
throw lastError
|
|
}
|
|
|
|
if (response.ok) return (await response.json()) as T
|
|
|
|
if (response.status === 401 && !credentialsInQuery) {
|
|
// Host may be stripping the Authorization header; the documented
|
|
// fallback is credentials in the query string (HTTPS enforced upstream).
|
|
credentialsInQuery = true
|
|
lastError = await parseError(response)
|
|
continue
|
|
}
|
|
if (RETRYABLE_STATUS.has(response.status) && attempt < RETRY_DELAYS_MS.length) {
|
|
lastError = await parseError(response)
|
|
await sleep(RETRY_DELAYS_MS[attempt])
|
|
continue
|
|
}
|
|
throw await parseError(response)
|
|
}
|
|
throw lastError instanceof Error
|
|
? lastError
|
|
: new WooCommerceApiError('WooCommerce request failed', 0)
|
|
}
|
|
|
|
export interface ListOrdersOptions {
|
|
/** ISO timestamp; interpreted as UTC (dates_are_gmt is always sent). */
|
|
modifiedAfter: string
|
|
page: number
|
|
}
|
|
|
|
/**
|
|
* One page of orders modified after the cursor, oldest-modified first so the
|
|
* caller's cursor advances chronologically. Requires WooCommerce 5.8+
|
|
* (modified_after); older stores fail with a woocommerce_rest_invalid_param
|
|
* style error surfaced to the connection's error state.
|
|
*/
|
|
export async function listOrdersPage(
|
|
creds: WooCredentials,
|
|
options: ListOrdersOptions,
|
|
): Promise<WooOrder[]> {
|
|
return wcGet<WooOrder[]>(creds, '/orders', {
|
|
modified_after: options.modifiedAfter,
|
|
dates_are_gmt: 'true',
|
|
status: 'any',
|
|
orderby: 'modified',
|
|
order: 'asc',
|
|
per_page: String(WC_PAGE_SIZE),
|
|
page: String(options.page),
|
|
})
|
|
}
|
|
|
|
/** Hard cap on refund pages per order; a real order never approaches this. */
|
|
const MAX_REFUND_PAGES = 10
|
|
|
|
/**
|
|
* All refunds of one order. Terminates on an EMPTY batch, not a short one
|
|
* (hosts may cap per_page below our request, same as the order pagination),
|
|
* dedupes by id so a host that ignores `page` cannot loop forever, and caps
|
|
* total pages as a final backstop.
|
|
*/
|
|
export async function listOrderRefunds(
|
|
creds: WooCredentials,
|
|
orderId: number,
|
|
): Promise<WooRefund[]> {
|
|
const refunds: WooRefund[] = []
|
|
const seen = new Set<number>()
|
|
for (let page = 1; page <= MAX_REFUND_PAGES; page++) {
|
|
const batch = await wcGet<WooRefund[]>(creds, `/orders/${orderId}/refunds`, {
|
|
per_page: String(WC_PAGE_SIZE),
|
|
page: String(page),
|
|
})
|
|
if (batch.length === 0) return refunds
|
|
const fresh = batch.filter((r) => !seen.has(r.id))
|
|
if (fresh.length === 0) return refunds
|
|
for (const refund of fresh) seen.add(refund.id)
|
|
refunds.push(...fresh)
|
|
}
|
|
// Cap exhausted with data still flowing: returning the partial list would
|
|
// let the sync advance its cursor past refunds it never saw. Throwing
|
|
// routes into the caller's refund-failure path instead (order held, cursor
|
|
// capped, retried next run).
|
|
throw new WooCommerceApiError(
|
|
`Refund pagination cap exceeded for order ${orderId}`,
|
|
0,
|
|
)
|
|
}
|
|
|
|
/**
|
|
* Verify credentials and read store metadata. The one-order probe is the
|
|
* authoritative credential check (it exercises the read scope the feed
|
|
* needs); title and settings lookups are best-effort extras.
|
|
*/
|
|
export async function testConnectionAndFetchStoreInfo(
|
|
creds: WooCredentials,
|
|
): Promise<WooStoreInfo> {
|
|
await wcGet<unknown[]>(creds, '/orders', { per_page: '1' })
|
|
|
|
const info: WooStoreInfo = {
|
|
name: null,
|
|
currency: null,
|
|
prices_include_tax: null,
|
|
wc_version: null,
|
|
}
|
|
|
|
try {
|
|
const settings = await wcGet<Array<{ id: string; value: unknown }>>(
|
|
creds,
|
|
'/settings/general',
|
|
)
|
|
const currency = settings.find((s) => s.id === 'woocommerce_currency')?.value
|
|
if (typeof currency === 'string' && currency) info.currency = currency.toUpperCase()
|
|
const pricesIncludeTax = settings.find((s) => s.id === 'woocommerce_prices_include_tax')?.value
|
|
if (typeof pricesIncludeTax === 'string') info.prices_include_tax = pricesIncludeTax === 'yes'
|
|
} catch {
|
|
// Settings need broader permissions on some setups; the feed works without.
|
|
}
|
|
|
|
try {
|
|
const status = await wcGet<{ environment?: { version?: string } }>(creds, '/system_status')
|
|
if (status.environment?.version) info.wc_version = status.environment.version
|
|
} catch {
|
|
// system_status is admin-capability data and often blocked; optional.
|
|
}
|
|
|
|
try {
|
|
// The WP REST index is public and carries the site title.
|
|
const response = await safeFetch(`${storeOriginOf(creds)}/wp-json/`, {
|
|
headers: { Accept: 'application/json' },
|
|
signal: AbortSignal.timeout(REQUEST_TIMEOUT_MS),
|
|
})
|
|
if (response.ok) {
|
|
const body = (await response.json()) as { name?: string }
|
|
if (body.name) info.name = body.name
|
|
}
|
|
} catch {
|
|
// Cosmetic only.
|
|
}
|
|
|
|
return info
|
|
}
|