# Compromised Credentials Credential-exposure findings surfaced by Island's dark-web / breach-corpus scanning: users on the tenant whose email + password pairs have appeared in public breach dumps, with breach source, status, and exposure timestamps. **Sync Type: Incremental** > Passwords themselves are never returned by Island — only the breach corpus name (`breachSource`) and exposure metadata. Monad never handles, stores, or forwards plaintext credentials. ## Prerequisites 1. An Island Enterprise Browser tenant with the Management Console accessible. 2. An account with the **admin** or **system-admin** role. 3. **The credential-exposure feature must be licensed on the tenant.** Not every Island tier includes dark-web scanning; a `403 Forbidden` here usually means the tenant lacks the entitlement. 4. Your tenant's hosting region (US or EU). ## Authentication Same credential as the sibling Devices and Users inputs — the **Management API** panel key, **not** the SIEM bearer token. ### The two-credential trap | Purpose | Console panel | | --- | --- | | **Management pull APIs (this input, Devices, Users)** | System Settings → Integrations → **API** | | SIEM event queue | System Settings → Integrations → **SIEM** → Generic SIEM Integration | **Do not paste the SIEM bearer token into the API Key field.** ## Configuration ### Settings | Setting | Type | Required | Description | | ------- | ---- | -------- | ----------- | | Region | string (enum) | Yes | Tenant hosting region. `us` → `management.island.io`, `eu` → `eu.management.island.io`. Match the domain you use to log into the Management Console. | | API Key | secret (string) | Yes | Management API key, from Console → Integrations → **API** panel. | | Backfill Start Time | string | No | ISO-8601 timestamp the first sync starts from (e.g. `2024-01-01T00:00:00Z`). Leave empty to start from now. Later syncs are incremental and ignore this value. | There is no cron setting on this input — it polls continuously. ## Backfill and Replay | Guarantee | Supported? | | --- | --- | | Historical backfill | Yes — set **Backfill Start Time** to pull findings dated from that point | | Replay | Partial — findings can repeat across syncs; dedupe downstream on `id` | | Incremental cursor | Yes — advances on each finding's `compromisedDate` | | `status` filtering | None applied — findings of every `status` are emitted; filter downstream | Note that `status` changes in Island (a finding moving to resolved) do not necessarily re-emit the record, since the cursor tracks `compromisedDate`. Treat this stream as "new exposures detected", not as a live status mirror. ## Rate limits and throughput Island publishes no explicit ceiling; Monad caps outbound calls at roughly **1 request/second** per input. The compromised-credentials corpus is typically far smaller than the devices or users inventories, so each sync is cheap. ## Troubleshooting - **`401 Unauthorized`** — Wrong credential type. Confirm the key is from the **API** integration panel, not the **SIEM** panel. - **`403 Forbidden`** — **Most commonly means the credential-exposure feature is not licensed on this tenant.** Also fires for non-admin accounts. - **No records emitted** — In this stream, usually good news. Confirm the feature is enabled and users are in scope for scanning. ## Example ```json { "id": "cc-48213-ae99d-71044", "breachSource": "Collection#1", "compromisedDate": "2026-09-11T07:01:51Z", "createdDate": "2026-09-11T07:05:00Z", "email": "user@example.com", "impactedDomain": "example.com", "status": "Unresolved", "tenantId": "tenant-001-acme", "updatedDate": "2026-09-11T09:17:23Z", "username": "art.rice204" } ``` ## Sync frequency By default this input polls approximately every 10 seconds, with each sync beginning after the previous one completes. A cron schedule configured on the pipeline overrides this cadence. See [Input Sync Frequency](/docs/guides/sync-frequency) for details.