> ## Documentation Index > Fetch the complete documentation index at: https://docs.ledgersyncappv2.com/llms.txt > Use this file to discover all available pages before exploring further. # Refresh a connection > Force a re-pull of accounts and recent transactions for this Connection. Banks usually post new transactions once a day so we already refresh active connections automatically — call this only when you have a specific reason to need fresher data (user just clicked a "refresh" button, you're reconciling something time-sensitive). ### Source-specific behavior On-demand refresh is not uniformly available across sources. The HTTP outcome depends on the connection's `source`: - **Finicity (`con_FINICITY_*`)** — returns `202 Accepted` with an `operation_id`. The backend queues the refresh asynchronously; on success the `account.refresh.completed` webhook fires when the bank call finishes, and on terminal failure the `account.refresh.failed` webhook fires instead. Refreshes **only the accounts already linked on this connection**: it updates balances and transactions on rows we already hold and never adds an account. An account the end-user opened, or chose not to share, after the original link is not picked up here no matter how often you call it. Use `POST /connections/{connection_id}/reauthorize` for that. - **MX (`con_MX_*`)** — returns `202 Accepted` with an `operation_id`. MX has no synchronous refresh API, so the backend triggers a member aggregation and returns immediately; MX pulls fresh balances and transactions asynchronously, and the `account.refresh.completed` webhook fires when it finishes (`account.refresh.failed` on terminal failure). The aggregation re-reads the member's **whole** account list, so an account added at the bank since link time is created here without any end-user step. The same happens on the scheduled passes, so on MX you can simply wait. - **FDE (`con_FDE_*`)** — returns `202 Accepted` with an `operation_id`. FDE re-runs its document extraction pipeline; on success the `account.refresh.completed` webhook fires on completion, and on terminal failure the `account.refresh.failed` webhook fires instead. The run re-reads the portal's sub-accounts, so a sub-account that has appeared since link time is created here too. ### Accounts opened after linking Only Finicity is closed: refresh there is update-only. MX and FDE discover new accounts on a refresh, and on MX the scheduled passes do the same without you calling anything. Either way nothing pushes you a notification, because there is no `account.added` webhook. After a refresh completes, re-list `GET /v3/accounts?client_id=…&connection_id=…` and diff on account id. `Account.created_at` carries when LedgerSync first saw the account, which is the field to sort by. ### Refresh support by source All three connectable sources — Finicity, MX, and FDE — support on-demand refresh and answer `202 Accepted` with an `operation_id`; the outcome always arrives asynchronously on the `account.refresh.completed` / `account.refresh.failed` webhook, never in the `202` body. Uploaded-statement (`PDF`) data is read-only, has no connection, and is never refreshed. ### Reading the operation The operation this returns reaches a terminal state, so polling `GET /operations/{operation_id}` is a real fallback when you cannot run a webhook handler. A refresh is one aggregation at the bank but reports per sub-account, so it stays `queued` until every account on the connection has reported, and resolves itself if those reports never arrive. On success `result` carries `accounts_refreshed` and an `accounts` array. A refresh where the aggregation ran but an account did not come back is reported as `failed` with `refresh_partially_failed`, carrying the same per-account breakdown under `error.accounts`, so branching on `status` alone cannot silently accept a connection that is missing an account's transactions. ### Rate limiting Refresh carries a daily per-customer budget separate from the standard request-rate tiers, because each call is a real aggregation at the institution rather than just another API request. Exceeding it returns `429` without queueing anything or contacting the bank. Active connections already refresh about once a day on their own, so this endpoint is for when something actually asked for fresher data. ### Empty-body POST This endpoint takes no request body. Send an explicit `Content-Length: 0` header — Google's HTTPS load balancer in front of the API rejects body-less POSTs without one with `411 Length Required`. Most HTTP clients add this automatically; some (raw `fetch`, certain SDKs in keep-alive mode) do not. ## OpenAPI ````yaml /openapi.yaml post /connections/{connection_id}/refresh openapi: 3.1.0 info: title: LedgerSync API version: '2026-05-22' summary: A modern REST API for connecting bank accounts and pulling financial data description: > > **First time here?** The [**Getting started** guide](https://portal.ledgersyncappv2.com/dashboard/getting-started) walks the full integration end-to-end (about 15 minutes). This page is the endpoint reference for after you've read it. Welcome. This is LedgerSync's API for connecting your users' bank accounts and pulling their transactions, statements, and account details. Everything is JSON over HTTPS. Errors are easy to read. Webhooks fire as state changes. Sandbox is one key away. ## Connection lifecycle and the two `connection.id` formats A Connection's `id` changes shape once the user finishes linking their bank — there are two distinct identifiers, and you must only use the canonical one to read accounts, transactions, or statements. 1. **Placeholder id (pending state).** `POST /v3/clients/{id}/connections` returns `202 Accepted`. The initiate operation first surfaces a `connection_with_action` result whose embedded `connection.id` is a UUID-prefixed placeholder of the form `con_` (e.g. `con_a1b2c3d4-e5f6-47a8-9b12-c3d4e5f6a7b8`), paired with a `widget_url`. The placeholder is **not** a valid id for any other endpoint — calling `GET /v3/connections/{placeholder_uuid}` returns `400 Bad Request` with `missing source separator`. Treat it as opaque routing state for the widget step only. 2. **Canonical id (active state).** Once the user finishes the widget and the source pushes its first callback, the connection moves to `status=active` and the `connection.active` webhook fires. Exactly two surfaces report the canonical id `con__` (for example `con_FINICITY_41294`, `con_MX_1224`): the `connection.active` webhook payload, and `GET /v3/clients/{id}/connections`. **This** is the id every other endpoint accepts (`/accounts`, `/transactions`, `/statements`, refresh, delete). **Tell the two apart by shape, not by where you got them.** The initiate operation's `result` is a snapshot written once, at the moment the widget URL is issued, and nothing rewrites it afterwards. The operation is already `succeeded` while the user still has the widget open, so `status=succeeded` does **not** mean the bank is linked, and re-polling later returns exactly what it returned the first time. For Finicity and MX that is a placeholder. (An FDE initiate can return the canonical id directly; checking the shape covers both without special-casing the source.) The correct integration pattern is therefore: - Initiate the connection, capture `operation_id`. - Open `result.connection.action.widget_url` for the user. Store `result.connection.id` only if it matches `con__`; never store a `con_`. - Wait for the `connection.active` webhook and read `data.connection.id` from it. If you cannot run a webhook handler, poll `GET /v3/clients/{id}/connections` and take the canonical id from there. Do not try to `GET /v3/connections/{placeholder_uuid}` during the pending window. Listing endpoints (`GET /v3/clients/{id}/connections`, `GET /v3/connections/{id}`) only ever return canonical ids, because they only surface connections that have reached `active`. ## Quick start Mint a sandbox key in the developer portal, then create your first **Client** (your end-user — the person whose bank we'll be reading): ```bash curl \ https://api-sandbox.ledgersyncappv2.com/v3/clients \ -H "Authorization: Bearer sk_test_..." \ -H "Content-Type: application/json" \ -d '{"email":"alice@example.com","name":"Alice"}' ``` You'll get back a `Client` object with an `id`. From there the full flow is: 1. **Register a webhook** at `POST /v3/webhooks/subscriptions` so you can be notified when the connection progresses. 2. **Initiate a connection** at `POST /v3/clients/{id}/connections` with an `institution_id` from `GET /v3/institutions`. LedgerSync's router picks the underlying source. Every source hands back a `widget_url` in the `connection.requires_action` result — for FDE it points at a LedgerSync-hosted connect page where credentials are entered, never sent to the API. 3. **Open the widget URL** in your user's browser. They pick their bank, log in, and choose accounts to share. 4. **Receive `connection.active`** on your webhook URL. List accounts and transactions. **Don't want to build your own bank picker?** Skip steps 2–3: call `POST /v3/clients/{id}/connect-session` to get a LedgerSync-hosted link, and email it to your member. They open it, search for their own bank, pick it, and connect it — all on a page we host, with no `institution_id` needed up front. You still receive `connection.active` on your webhook exactly as above. Revoke a link at any time with `DELETE /v3/clients/{id}/connect-session/{sid}`. (This is the v3 replacement for the old `account/add/lite` widget.) Want a step-by-step walkthrough with curl per step plus a sandbox shortcut that skips the widget? Read the [Getting started guide](https://portal.ledgersyncappv2.com/dashboard/getting-started). ## Authentication Pass your secret key in the `Authorization` header as a Bearer token: `Authorization: Bearer sk_test_...` (sandbox) or `Bearer sk_live_...` (production). Treat secret keys like passwords — never embed them in mobile apps or front-end code. ## Conventions **Sync vs async.** Most endpoints respond synchronously — you get the resource back right away. A handful of flows are genuinely async (initiating a bank connection, extracting a statement, generating a verification report); those return `202 Accepted` with an `operation_id` you can poll, and the matching webhook fires when the work finishes. **Errors.** Every error is `{ "error": { "code", "message", "doc_url", "type" } }`. Branch on `code`. Click `doc_url` for the troubleshooting page. Every response carries an `X-LS-Trace-Id` header — paste it in support tickets and we can jump straight to your request. **Webhooks.** Every delivery carries `X-LS-Webhook-Signature`: hex HMAC-SHA256 over `X-LS-Webhook-Timestamp`, a literal `.`, and the **raw request body**, in that order. Signing the body alone will not match. Verify before trusting the payload. Full event catalog + signature example on the [Webhooks tab](#webhooks). ## Need help? Email [support@ledgersync.com](mailto:support@ledgersync.com) or open a thread in the developer portal. Quote the `X-LS-Trace-Id` from your response — it makes everything faster. contact: name: LedgerSync Developer Support email: support@ledgersync.com url: https://portal.ledgersyncappv2.com license: name: Proprietary url: https://ledgersync.com/terms servers: - url: https://api-sandbox.ledgersyncappv2.com/v3 description: >- Sandbox — use `sk_test_...` keys from the developer portal. Routes to the real Finicity and MX sandbox banks (FinBank, mxbank) via the same connector code paths as production. - url: https://api.ledgersyncappv2.com/v3 description: >- Production — use `sk_live_...` keys. Hits real banks via Finicity, MX, or FDE depending on the connection. security: - bearer: [] tags: - name: Clients description: | A **Client** is the end-user whose bank accounts you're managing — usually a real person or business. You create one Client per user before linking any accounts. Pass your own `external_id` to join Clients back to records in your system. - name: Connections description: | A **Connection** links a Client to a bank or financial-data source (Finicity, MX, or FDE). One Client can have many Connections — one per bank they've linked. You initiate a Connection, the user finishes it (widget or MFA), and a `connection.active` webhook tells you when it's ready. - name: Accounts description: | Once a Connection goes active, the bank's accounts (checking, savings, credit, loans) show up here. Use these endpoints to list, inspect, and refresh them. - name: Transactions description: | Posted transactions and pending charges for an account. New transactions are surfaced by a future push-delivery event for transactions after each refresh; you can also list/page them directly. - name: Statements description: | Period statements (usually monthly PDFs) the bank publishes. For FDE-sourced statements, extracted line items are attached after OCR completes. - name: Checks description: | Images of paper checks, captured alongside FDE statement extraction. FDE-only: Finicity and MX report `check_images` as `unsupported`. Fronts only — LedgerSync does not capture the back of a check, so there is no `side` field. - name: Operations description: | The handful of LedgerSync calls that genuinely run asynchronously return an `operation_id`. Poll it here, or just listen for the matching webhook — your call. - name: Webhooks description: | Manage where LedgerSync delivers event notifications. Each subscription has its own HMAC signing secret. Rotate it whenever you want — the old secret stays valid for 24 hours so deploys don't break verification. - name: ApiKeys description: | Create, list, and revoke API keys. Keys are scoped to one environment (sandbox or live). The plaintext secret is returned exactly once at creation — store it somewhere safe. - name: Institutions description: | The unified v3 institution catalog. Search by name; pick a row; pass its `id` to `POST /clients/{client_id}/connections`. LedgerSync's router decides which underlying source (Finicity, MX) to use — integrators don't pick a source. - name: Settings description: | Account-level settings the integrator manages once per environment. Currently exposes the redirect-URL allowlist consulted at `/connections/initiate`. Same scope as Stripe Connect / Plaid Link / OAuth — register your app's redirect URIs once, not per end-user. - name: Sandbox description: | Sandbox-only helpers. Use `GET /sandbox/institutions` to list the sandbox-eligible test banks you can target when creating a sandbox connection (Finicity FinBank, MX `mxbank`, FDE test extractors). Lifecycle transitions and webhooks come from the same real connector paths as live traffic — no synthetic lifecycle driver, no replay fixtures. Drive the widget with the public test bank credentials documented in the developer portal. - name: Metrics description: | Read-only aggregations powering the developer portal's "Metrics" page. Every endpoint is scoped to the authenticated principal's customer and environment — `sandbox` and `live` data never mix across the wire. Default window when `from`/`to` are omitted is the last 30 days (max 366). - name: Health description: A simple liveness probe. No auth required. - name: PortalGating description: | Developer-portal access-request endpoints (sandbox + live gating). These are HMAC-signed calls from the LedgerSync portal to v3 and not part of the integrator-facing API surface. Documented here so the single spec stays authoritative. paths: /connections/{connection_id}/refresh: parameters: - in: path name: connection_id required: true description: | Canonical Connection id (`con__`). The placeholder UUID returned during initiate is rejected here with 400 — only `active` connections can be refreshed. schema: type: string example: con_FINICITY_41294 post: tags: - Connections summary: Refresh a connection description: | Force a re-pull of accounts and recent transactions for this Connection. Banks usually post new transactions once a day so we already refresh active connections automatically — call this only when you have a specific reason to need fresher data (user just clicked a "refresh" button, you're reconciling something time-sensitive). ### Source-specific behavior On-demand refresh is not uniformly available across sources. The HTTP outcome depends on the connection's `source`: - **Finicity (`con_FINICITY_*`)** — returns `202 Accepted` with an `operation_id`. The backend queues the refresh asynchronously; on success the `account.refresh.completed` webhook fires when the bank call finishes, and on terminal failure the `account.refresh.failed` webhook fires instead. Refreshes **only the accounts already linked on this connection**: it updates balances and transactions on rows we already hold and never adds an account. An account the end-user opened, or chose not to share, after the original link is not picked up here no matter how often you call it. Use `POST /connections/{connection_id}/reauthorize` for that. - **MX (`con_MX_*`)** — returns `202 Accepted` with an `operation_id`. MX has no synchronous refresh API, so the backend triggers a member aggregation and returns immediately; MX pulls fresh balances and transactions asynchronously, and the `account.refresh.completed` webhook fires when it finishes (`account.refresh.failed` on terminal failure). The aggregation re-reads the member's **whole** account list, so an account added at the bank since link time is created here without any end-user step. The same happens on the scheduled passes, so on MX you can simply wait. - **FDE (`con_FDE_*`)** — returns `202 Accepted` with an `operation_id`. FDE re-runs its document extraction pipeline; on success the `account.refresh.completed` webhook fires on completion, and on terminal failure the `account.refresh.failed` webhook fires instead. The run re-reads the portal's sub-accounts, so a sub-account that has appeared since link time is created here too. ### Accounts opened after linking Only Finicity is closed: refresh there is update-only. MX and FDE discover new accounts on a refresh, and on MX the scheduled passes do the same without you calling anything. Either way nothing pushes you a notification, because there is no `account.added` webhook. After a refresh completes, re-list `GET /v3/accounts?client_id=…&connection_id=…` and diff on account id. `Account.created_at` carries when LedgerSync first saw the account, which is the field to sort by. ### Refresh support by source All three connectable sources — Finicity, MX, and FDE — support on-demand refresh and answer `202 Accepted` with an `operation_id`; the outcome always arrives asynchronously on the `account.refresh.completed` / `account.refresh.failed` webhook, never in the `202` body. Uploaded-statement (`PDF`) data is read-only, has no connection, and is never refreshed. ### Reading the operation The operation this returns reaches a terminal state, so polling `GET /operations/{operation_id}` is a real fallback when you cannot run a webhook handler. A refresh is one aggregation at the bank but reports per sub-account, so it stays `queued` until every account on the connection has reported, and resolves itself if those reports never arrive. On success `result` carries `accounts_refreshed` and an `accounts` array. A refresh where the aggregation ran but an account did not come back is reported as `failed` with `refresh_partially_failed`, carrying the same per-account breakdown under `error.accounts`, so branching on `status` alone cannot silently accept a connection that is missing an account's transactions. ### Rate limiting Refresh carries a daily per-customer budget separate from the standard request-rate tiers, because each call is a real aggregation at the institution rather than just another API request. Exceeding it returns `429` without queueing anything or contacting the bank. Active connections already refresh about once a day on their own, so this endpoint is for when something actually asked for fresher data. ### Empty-body POST This endpoint takes no request body. Send an explicit `Content-Length: 0` header — Google's HTTPS load balancer in front of the API rejects body-less POSTs without one with `411 Length Required`. Most HTTP clients add this automatically; some (raw `fetch`, certain SDKs in keep-alive mode) do not. operationId: refreshConnection parameters: - $ref: '#/components/parameters/IdempotencyKey' requestBody: required: false description: | No body. Send `Content-Length: 0` so the upstream load balancer doesn't reject the request with 411. content: application/json: schema: type: object nullable: true responses: '202': $ref: '#/components/responses/Accepted' '401': $ref: '#/components/responses/Unauthorized' '404': $ref: '#/components/responses/NotFound' '411': description: | Returned by the upstream HTTPS load balancer when a POST arrives without a `Content-Length` header. Retry with `Content-Length: 0` set explicitly. '429': $ref: '#/components/responses/TooManyRequests' '502': description: | The upstream aggregator or the backend was unreachable while starting the refresh, so nothing was queued. Body carries the canonical `ErrorEnvelope` (typically `error.code = "upstream_unavailable"`); retry shortly. headers: X-LS-Trace-Id: $ref: '#/components/headers/X-LS-Trace-Id' content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' components: parameters: IdempotencyKey: name: Idempotency-Key in: header required: false description: | > **Reserved, not yet honored.** The header is accepted and > ignored today: nothing reads it, no response is replayed, > and `idempotency_conflict` is never returned. **Do not > auto-retry a failed POST on the assumption that this > protects you** — a retried `POST /clients` or > `POST /clients/{id}/connections` creates a duplicate. Send > the header if you want to be ready for it; do not depend on > it. The behavior described below is the intended contract > for a future release. Safe-retry key for POSTs. Send any unique string per logical request (a UUIDv4 is great). If the network drops and you retry with the same `Idempotency-Key` within 24 hours, you get the exact same response back instead of creating a duplicate. **Cached responses include failures (4xx and 5xx) too.** If a call failed because of a bad input and you want to try again with the corrected input, use a **fresh** key — otherwise you'll keep getting the cached failure. schema: type: string maxLength: 255 example: 6f1a8c50-3e9c-4d4a-b1f5-2c5b9a2f7d11 responses: Accepted: description: | The request was accepted but the underlying work runs asynchronously. Use the returned `operation_id` to poll `GET /operations/{operation_id}` or just wait for the matching webhook event. headers: X-LS-Trace-Id: $ref: '#/components/headers/X-LS-Trace-Id' content: application/json: schema: $ref: '#/components/schemas/OperationAck' Unauthorized: description: | Either the `Authorization` header is missing or the bearer token doesn't match an active API key. Double-check the key and the environment — sandbox keys can't be used against the production base URL and vice versa. headers: X-LS-Trace-Id: $ref: '#/components/headers/X-LS-Trace-Id' content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' NotFound: description: | We didn't find a resource matching the path parameters. Common causes: the id belongs to a different customer's key, the resource was deleted, or there's a typo in the id. headers: X-LS-Trace-Id: $ref: '#/components/headers/X-LS-Trace-Id' content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' TooManyRequests: description: | You hit the rate limit. Back off, then retry after the number of seconds in the `Retry-After` header. The `X-RateLimit-Remaining` and `X-RateLimit-Reset` headers tell you the current window state. headers: X-LS-Trace-Id: $ref: '#/components/headers/X-LS-Trace-Id' X-RateLimit-Limit: $ref: '#/components/headers/X-RateLimit-Limit' X-RateLimit-Remaining: $ref: '#/components/headers/X-RateLimit-Remaining' X-RateLimit-Reset: $ref: '#/components/headers/X-RateLimit-Reset' X-RateLimit-Policy: $ref: '#/components/headers/X-RateLimit-Policy' Retry-After: $ref: '#/components/headers/Retry-After' content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' headers: X-LS-Trace-Id: description: Distributed-trace ID for this request. Quote in support tickets. schema: type: string example: 4bf92f3577b34da6a3ce929d0e0e4736 X-RateLimit-Limit: description: | Requests allowed in the per-minute window. The day-window and the live-tier burst (2x over 60s) are advertised via `X-RateLimit-Policy`. Per-environment numbers: - sandbox: 120 req/min, 10,000 req/day - live: 60 req/min, 50,000 req/day, 2x burst over 60s Connection-initiation concurrency (sandbox 10 / live 50) and webhook-event/day budgets (sandbox 5k / live 25k) are enforced at the service layer and are NOT counted in these headers. schema: type: integer X-RateLimit-Remaining: description: Lowest remaining count across all active windows. schema: type: integer X-RateLimit-Reset: description: Unix seconds when the per-minute window resets. schema: type: integer X-RateLimit-Policy: description: | IETF draft-ietf-httpapi-ratelimit-headers — comma-separated policy entries `;w=` advertising every enforced window so an SDK consumer can tell which window tripped them. Example: `60;w=60, 50000;w=86400`. schema: type: string Retry-After: description: Seconds to wait before retrying (on 429 / 503). schema: type: integer schemas: ErrorEnvelope: type: object description: | Error envelope returned as the top-level body of every non-2xx HTTP response. The actual error payload lives in the `error` field — branch on `error.code`. required: - error properties: error: $ref: '#/components/schemas/ErrorInfo' OperationAck: type: object description: Returned with 202 Accepted for async operations. required: - operation_id - status - poll_url properties: operation_id: type: string example: op_01HXYZ8A6N7K2W9PQ4T5Z3V6E0 status: type: string enum: - queued poll_url: type: string format: uri example: >- https://api-sandbox.ledgersyncappv2.com/v3/operations/op_01HXYZ8A6N7K2W9PQ4T5Z3V6E0 estimated_seconds: type: integer description: Rough ETA for completion. Best-effort. ErrorInfo: type: object description: | The inner error payload. Used as the top-level body of an HTTP error response (wrapped in `ErrorEnvelope`) and as the embedded `error` field on resources that record a failure (like `Operation.error`). required: - code - message - type properties: code: type: string description: | Stable machine-readable identifier. Branch on this in your code, not on `message`. example: unknown_api_key message: type: string description: Plain-English explanation safe to log. example: The API key you presented doesn't match any active key. doc_url: type: string format: uri description: | Link to the docs page for this specific error code. Shareable with teammates in support tickets. example: https://portal.ledgersyncappv2.com/errors/unknown_api_key type: type: string description: | Stripe-style broad failure-mode category — useful for "treat all of these the same way" branches. Derived from the HTTP status. Orthogonal to `category`, which classifies by origin. enum: - auth_error - invalid_request - rate_limit_error - idempotency_error - not_found - api_error example: auth_error category: type: string description: | Plaid-style coarse classification by origin. Branch on this for routing logic — retry the request, surface to the end user in the widget, or page on-call. Orthogonal to `type` (which is HTTP-status-derived). enum: - AUTH_ERROR - INSTITUTION_ERROR - CAPABILITY_UNAVAILABLE - CONNECTION_ERROR - RATE_LIMIT - INVALID_REQUEST - RESOURCE_NOT_FOUND - PLATFORM_ERROR example: AUTH_ERROR is_user_actionable: type: boolean description: | True when the end user can resolve this error (re-enter credentials, complete MFA, accept an updated agreement). False when the error requires institution-side or LS platform-side action. Useful for deciding whether to send the end user back to the widget or surface an "operational issue" banner. example: true source_diagnostic_code: type: string pattern: ^(FIN|MX|FDE)-[A-Z0-9_]+$ description: | Opaque upstream-source diagnostic identifier (e.g. `FIN-103`, `MX-DENIED`). Present only on errors that originated at an aggregator/extractor. Use for support triage — do NOT branch on this value; the unified `code` and `category` are the integrator-facing vocabulary. example: FIN-103 param: type: string description: | On every 400 or 413 that is about one request field (`validation_failed` and `invalid_request` alike), the field that tripped the check. Dot-separated path for nested fields. Absent on errors that are not about a field (auth, not found, upstream faults). example: client.email trace_id: type: string description: | Same as the `X-LS-Trace-Id` response header — paste in a support ticket to jump to the request. example: 4bf92f3577b34da6a3ce929d0e0e4736 errors: type: array description: | Per-field validation errors (only present on 400 when multiple fields failed at once). items: type: object required: - param - message properties: param: type: string example: client.email message: type: string example: must be a valid email address code: type: string example: invalid_email securitySchemes: bearer: type: http scheme: bearer bearerFormat: LedgerSync API key description: | Pass your secret key in the `Authorization` header as a Bearer token: `Authorization: Bearer sk_test_...` (sandbox) or `Bearer sk_live_...` (production). Keys are created in the developer portal and the plaintext secret is shown exactly once at creation. Treat them like passwords — never embed them in mobile apps or front-end code. ````