# List API request logs API Reference Returns the history of public API requests made by you, with any of your API keys in this environment — not only the key you are authenticating with. Only `auth_type=API_KEY` traffic is recorded. - **The axis is the person, not the individual credential:** a second key of yours sees the same history, and narrowing it to one key is a filter (`api_key_id`), not the default. - **It is still not the account's traffic:** requests made by other users of the same account, or by their API keys, are never returned. The `{account_id}` in the path authorizes the call; it does not widen what you can see. - **Environment is not a filter:** results are always scoped to the environment of the credential you authenticate with — a `beel_sk_test_*` key sees the test traffic of all your test keys, a `beel_sk_live_*` key the live traffic of all your live ones. To see the other environment, use a key from that environment. - **Cursor pagination:** navigate with the opaque `cursor` returned in `next_cursor` / `prev_cursor`; there is no jump to an arbitrary page N. - **Time window:** defaults to the last 30 days; narrow or move it with `from`/`to`. ## GET /v1/accounts/{account_id}/request-logs **List API request logs** Returns the history of public API requests made by you, with any of your API keys in this environment — not only the key you are authenticating with. Only `auth_type=API_KEY` traffic is recorded. - **The axis is the person, not the individual credential:** a second key of yours sees the same history, and narrowing it to one key is a filter (`api_key_id`), not the default. - **It is still not the account's traffic:** requests made by other users of the same account, or by their API keys, are never returned. The `{account_id}` in the path authorizes the call; it does not widen what you can see. - **Environment is not a filter:** results are always scoped to the environment of the credential you authenticate with — a `beel_sk_test_*` key sees the test traffic of all your test keys, a `beel_sk_live_*` key the live traffic of all your live ones. To see the other environment, use a key from that environment. - **Cursor pagination:** navigate with the opaque `cursor` returned in `next_cursor` / `prev_cursor`; there is no jump to an arbitrary page N. - **Time window:** defaults to the last 30 days; narrow or move it with `from`/`to`. ### Authentication Accepts any of: - `ApiKeyAuth` (HTTP bearer, token format `beel_sk_*`) ### Parameters - **account_id** (required) in path `string`: Account the call is authorized against. It does not widen the result set. - **only_errors** (optional) in query `boolean` (default: false): If true, only requests with status >= 400. - **method** (optional) in query `string`: Filter by HTTP method. - **http_status** (optional) in query `integer`: Filter by an exact HTTP status code. - **path_contains** (optional) in query `string`: Case-insensitive substring of the actual request path, IDs included (for example `/v1/companies/3f1c9a2e-5b7d-4e8f-9a1b-2c3d4e5f6a7b/invoices/8a2b4c6d-1e3f-4a5b-8c7d-9e0f1a2b3c4d/send`), so a path template with `{placeholders}` never matches, and a full path only matches requests made to that exact route shape. Prefer a resource segment such as `/invoices`, which matches that resource on every route that serves it. - **api_key_id** (optional) in query `string`: Narrow the result to one of your API keys. Any key of yours in this environment is accepted, not just the one you authenticate with; a key belonging to someone else simply yields no results. - **from** (optional) in query `string`: Lower bound of the time range (inclusive). Defaults to 30 days ago. - **to** (optional) in query `string`: Upper bound of the time range (inclusive). Defaults to now. - **cursor** (optional) in query `string`: Opaque cursor returned by a previous response (next_cursor / prev_cursor). - **limit** (optional) in query `integer` (default: 25): Maximum number of log entries in the page, from 1 to 100. Defaults to 25. ### Responses #### 200: Logs retrieved successfully. **Content `application/json`:** - **success** (required) `boolean`: No description (example: true) - **data** (required) `object`: No description - **request_logs** (required) `array[RequestLogSummary]`: No description - **pagination** `RequestLogCursorPagination`: Cursor-based pagination (keyset on _time + request_id). APL does not support cheap offset. - **meta** `ResponseMeta` #### 400: Bad request. From a dashboard session, a `BeeL-Environment` header whose value is not a known environment answers `ENVIRONMENT_HEADER_INVALID`, with the header named in `error.details` (`{"BeeL-Environment": "invalid"}`); it is never read as production. `BUSINESS_RULE_VIOLATION` is the generic answer to an operation the resource does not admit in its current state when that state has no code of its own. It carries no `details`; read the resource again before retrying. **Content `application/json`:** - **success** (required) `boolean`: No description (example: false) - **error** (required) `ErrorDetail` - **meta** `ResponseMeta` - **type** `string` (uri): Stable URI that identifies the problem type and where the integrator will find its documentation. (example: "https://docs.beel.es/errors/INVOICE_NO_LINES") - **title** `string`: Short summary of the problem type. Stable between occurrences of the same `type`. (example: "INVOICE_NO_LINES") - **detail** `string`: Specific message for this occurrence, localized according to `Accept-Language` (`es`, `en`, `ca`); defaults to English. Matches the legacy `error.message`. (example: "The invoice must have at least one line") - **instance** `string` (uri-reference): URI that identifies the specific occurrence of the problem — typically the path of the affected resource. (example: "/v1/invoices/abc-123") **Example:** ```json { "success": false, "error": { "code": "BAD_REQUEST", "message": "Invalid request" }, "meta": { "timestamp": "2025-01-15T10:30:00Z", "request_id": "4bf92f3577b34da6a3ce929d0e0e4736" } } ``` #### 401: Missing or invalid authentication. Like every other error, `message`/`detail` is localized according to `Accept-Language` (`es`, `en`, `ca`); defaults to English when the header is missing or asks for none of those. **Content `application/json`:** - **success** (required) `boolean`: No description (example: false) - **error** (required) `ErrorDetail` - **meta** `ResponseMeta` - **type** `string` (uri): Stable URI that identifies the problem type and where the integrator will find its documentation. (example: "https://docs.beel.es/errors/INVOICE_NO_LINES") - **title** `string`: Short summary of the problem type. Stable between occurrences of the same `type`. (example: "INVOICE_NO_LINES") - **detail** `string`: Specific message for this occurrence, localized according to `Accept-Language` (`es`, `en`, `ca`); defaults to English. Matches the legacy `error.message`. (example: "The invoice must have at least one line") - **instance** `string` (uri-reference): URI that identifies the specific occurrence of the problem — typically the path of the affected resource. (example: "/v1/invoices/abc-123") **Example:** ```json { "success": false, "error": { "code": "UNAUTHORIZED", "message": "Authentication is required to access this resource" }, "meta": { "timestamp": "2025-01-15T10:30:00Z", "request_id": "4bf92f3577b34da6a3ce929d0e0e4736" } } ``` #### 403: Authenticated but not allowed. Fourteen causes shared by every operation, told apart by `error.code`; an operation that adds a cause of its own names it in its `403`. You can branch on them exhaustively. - `INSUFFICIENT_SCOPE` — the credential lacks a scope the operation requires; `error.details.missing_scopes` names them as a single comma-separated string (for example `"invoices:write,customers:read"`), not as an array; `required_scopes` has the same shape and lists every scope the operation needs. Retrying will not help: mint a key that holds them. - `COMPANY_READ_ONLY` — the scope is there, but your access level over that NIF only lets you read it. - `ACCOUNT_MANAGEMENT_FORBIDDEN` — the scope is there, but your role over the account, or the management relationship you hold over it, does not cover this operation. - `ACCOUNT_NOT_ACCESSIBLE` — the account is not yours to reach, which is also the answer when it does not exist, so existence is never disclosed. - `ACTIVE_COMPANY_NOT_ACCESSIBLE` — the same, for a NIF: the company in the path, or the one named by `BeeL-Active-Company`, is not one this credential may reach — and again, this is also the answer when it does not exist. - `COMPANY_ACCESS_REVOKED` — your access to the NIF was withdrawn while the request was already in flight, so the write was rejected and nothing was recorded. Retrying will not help until the access is granted again. - `LIVE_CREDENTIAL_REQUIRED` — the operation changes the real account and the call came from a test API key (`beel_sk_test_…`). Use your live key or the dashboard. - `FEATURE_NOT_AVAILABLE` — your subscription does not include the entitlement the operation needs; `error.details.feature_code` names it. This gate runs **before** the scope gate, so for such an operation you never see `INSUFFICIENT_SCOPE` first. - `NO_ACTIVE_ACCOUNT` — the credential does not resolve to an account, so no scope can be evaluated against one. Fail-closed, not a permission that can be granted to you. - `OPERATION_REQUIRES_SESSION` — the operation is available only from the web session; no API key and no OAuth2 token can perform it, whatever scopes it holds. - `OPERATION_REQUIRES_API_KEY` — the reverse: the operation needs a scope that only an API key can hold (such as `accounts:write`), so a dashboard session never satisfies it. - `API_KEY_LIVE_NOT_AVAILABLE` — the call came with a live API key (`beel_sk_live_…`) and the account has no Live mode yet. It is never degraded to Test: use your test key, or activate Live first. - `ACCOUNT_REQUIRES_FOCUS_SWITCH` — dashboard session only: you are a member of the account in the path, but it is not the account in focus. Switch focus to it and retry. - `REQUEST_BLOCKED` — the edge security filter stopped the request before the API saw its body, usually because a text field carries something that looks like markup or a script (HTML tags, `