@celigo/api-specs 0.2.167 → 0.2.169

package/dist/api.yml

@@ -5932,6 +5932,14 @@ components:
             Do NOT include the protocol prefix (e.g., use "ftp.example.com", not "sftp://ftp.example.com").
           examples:
             - sftp.example.com
+          x-celigo-ai-guidance:
+            - |-
+              Don't point a connection at this environment's own Celigo file server (its
+              `endpoint.hostname`, with a username ending in `@<identifier>`, the server's own
+              `identifier`): a flow reads those files from Celigo Storage directly, and the save
+              refuses that pairing with 422 `fileserver_loopback_not_supported`. Another
+              environment's file server on the same host, such as a trading partner's, is an
+              ordinary connection.
         port:
           type: number
           description: |
@@ -6055,7 +6063,8 @@ components:
           description: |
             Maximum number of concurrent file transfer operations.
             FTP servers often have low connection limits — keep this value conservative.
-            Send no more than the account's licensed maximum (25 standard, 50 with an Environments license).
+            Send no more than the account's licensed maximum, the license's `concurrency` (25 standard,
+            50 with an Environments license).
             Unlike the rest of the `ftp` object, a `PUT` that omits this field keeps the stored value instead of clearing it.
           minimum: 1
           maximum: 50
@@ -6079,7 +6088,7 @@ components:
           description: |-
             Number of parallel transfer threads for this FTP connection. Must be a
             whole number — decimals are rejected on save. Send no more than the
-            account's licensed concurrency maximum. On
+            account's licensed concurrency maximum, the same as for `concurrencyLevel`. On
             save the value is mirrored into `userConcurrencyLevel` (and into
             `userTargetConcurrencyLevel` when auto-recovery of rate-limit errors
             is active with `targetConcurrencyLevel` set); clearing it clears the

package/dist/connection.yml

@@ -1185,7 +1185,8 @@ paths:
         `POST /v1/connections/{_id}/ping`, with no request body to send. Runs the adaptor's ping
         routine against the connection identified by `_id` to verify its credentials and network
         configuration still work; no data is read, written, or modified. An `http` connection with
-        no `http.ping` answers `{code: 200}` without sending any request.
+        no `http.ping` answers `{code: 200}` without sending any request, unless it authorizes with
+        an authorization code and holds no token yet: that one answers `missing_token`.
 
         Inspect the response body, not the HTTP status, to determine the outcome: success and
         adaptor-level failures both return HTTP 200 — success is `{code: 200}`, failures are
@@ -1259,8 +1260,10 @@ paths:
           whichever verb the client supports. Both verbs return HTTP 200 with {code: 200} on
           success and HTTP 200 with {errors: [...]} on adaptor-level failure, so inspect the body,
           not the status code. On an `http` connection, {code: 200} shows reachability only once
-          `http.ping` is set: with no ping configured, nothing is sent and the answer is always
-          {code: 200}.
+          `http.ping` is set: with no ping configured, nothing is sent and the answer is
+          {code: 200}, or `missing_token` for an authorization-code OAuth connection that holds no
+          token yet. An OAuth connection requests a token only when the target answers 401, so a
+          ping of an endpoint that needs no authentication passes without testing the OAuth setup.
     post:
       x-internal: false
       summary: Ping an existing connection
@@ -1270,8 +1273,11 @@ paths:
         adaptor-specific (HTTP sends the `http.ping` request, RDBMS/JDBC open a driver session,
         MCP issues a tools/list, NetSuite/Salesforce do a token or session validation). An `http`
         connection with no `http.ping` passes with `{code: 200}` without sending any request, so
-        the ping proves nothing about it until a ping is configured. The request body is optional
-        and ignored.
+        the ping proves nothing about it until a ping is configured; one that authorizes with an
+        authorization code and holds no token yet answers `missing_token` instead. An OAuth
+        connection requests a token only when the target answers 401, so a ping of an endpoint
+        that needs no authentication passes without testing the OAuth setup. The request body
+        is optional and ignored.
 
         Inspect the response body, not the HTTP status, to determine the outcome: success and
         adaptor-level failures both return HTTP 200 — success is `{code: 200}`, failures are
@@ -1383,8 +1389,10 @@ paths:
           - Return connection status and any error details
 
           Exception: an `http` connection with no `http.ping` answers {code: 200} without sending
-          any request, so it proves nothing until a ping (ideally one that needs the connection's
-          credentials) is configured.
+          any request (or `missing_token`, for an authorization-code OAuth connection that holds no
+          token yet), so it proves nothing until a ping is configured — one that needs the
+          connection's credentials, since an OAuth connection requests a token only when the target
+          answers 401.
   /v1/connections/{_id}/proxy:
     post:
       x-internal: false
@@ -2942,6 +2950,8 @@ paths:
         `/v1/connections/{_id}/oauth2` is an alias that answers identically. The URL's
         `state` token is one-time and expires quickly, so request a fresh one if the flow is
         interrupted, then poll `GET /v1/connections/{_id}` until `needsAuthorization` is `false`.
+        On a connection that uses PKCE (an iClient with `oauth2.pkceCodeChallengeMethod`), each call
+        issues a new `code_challenge`, so send the user to the most recent URL only.
 
         The URL's `redirect_uri` is the account's Celigo callback
         (`https://integrator.io/connection/oauth2callback` on North America), which the provider's
@@ -4780,6 +4790,14 @@ components:
             Do NOT include the protocol prefix (e.g., use "ftp.example.com", not "sftp://ftp.example.com").
           examples:
             - sftp.example.com
+          x-celigo-ai-guidance:
+            - |-
+              Don't point a connection at this environment's own Celigo file server (its
+              `endpoint.hostname`, with a username ending in `@<identifier>`, the server's own
+              `identifier`): a flow reads those files from Celigo Storage directly, and the save
+              refuses that pairing with 422 `fileserver_loopback_not_supported`. Another
+              environment's file server on the same host, such as a trading partner's, is an
+              ordinary connection.
         port:
           type: number
           description: |
@@ -4903,7 +4921,8 @@ components:
           description: |
             Maximum number of concurrent file transfer operations.
             FTP servers often have low connection limits — keep this value conservative.
-            Send no more than the account's licensed maximum (25 standard, 50 with an Environments license).
+            Send no more than the account's licensed maximum, the license's `concurrency` (25 standard,
+            50 with an Environments license).
             Unlike the rest of the `ftp` object, a `PUT` that omits this field keeps the stored value instead of clearing it.
           minimum: 1
           maximum: 50
@@ -4927,7 +4946,7 @@ components:
           description: |-
             Number of parallel transfer threads for this FTP connection. Must be a
             whole number — decimals are rejected on save. Send no more than the
-            account's licensed concurrency maximum. On
+            account's licensed concurrency maximum, the same as for `concurrencyLevel`. On
             save the value is mirrored into `userConcurrencyLevel` (and into
             `userTargetConcurrencyLevel` when auto-recovery of rate-limit errors
             is active with `targetConcurrencyLevel` set); clearing it clears the

package/dist/enduser.yml

@@ -2847,6 +2847,14 @@ components:
             Do NOT include the protocol prefix (e.g., use "ftp.example.com", not "sftp://ftp.example.com").
           examples:
             - sftp.example.com
+          x-celigo-ai-guidance:
+            - |-
+              Don't point a connection at this environment's own Celigo file server (its
+              `endpoint.hostname`, with a username ending in `@<identifier>`, the server's own
+              `identifier`): a flow reads those files from Celigo Storage directly, and the save
+              refuses that pairing with 422 `fileserver_loopback_not_supported`. Another
+              environment's file server on the same host, such as a trading partner's, is an
+              ordinary connection.
         port:
           type: number
           description: |
@@ -2970,7 +2978,8 @@ components:
           description: |
             Maximum number of concurrent file transfer operations.
             FTP servers often have low connection limits — keep this value conservative.
-            Send no more than the account's licensed maximum (25 standard, 50 with an Environments license).
+            Send no more than the account's licensed maximum, the license's `concurrency` (25 standard,
+            50 with an Environments license).
             Unlike the rest of the `ftp` object, a `PUT` that omits this field keeps the stored value instead of clearing it.
           minimum: 1
           maximum: 50
@@ -2994,7 +3003,7 @@ components:
           description: |-
             Number of parallel transfer threads for this FTP connection. Must be a
             whole number — decimals are rejected on save. Send no more than the
-            account's licensed concurrency maximum. On
+            account's licensed concurrency maximum, the same as for `concurrencyLevel`. On
             save the value is mirrored into `userConcurrencyLevel` (and into
             `userTargetConcurrencyLevel` when auto-recovery of rate-limit errors
             is active with `targetConcurrencyLevel` set); clearing it clears the

package/dist/eventreport.yml

@@ -43,8 +43,9 @@ paths:
         `startTime` must be within the last 30 days and the span to `endTime` must not exceed 3 days.
         If `endTime` is omitted, the server defaults it to the current time. Poll
         `GET /v1/eventreports/{_id}` until `status` is `completed`, then download the CSV via
-        `GET /v1/eventreports/{_id}/signedURL`. The create response includes `_requestedByUserId`;
-        subsequent GET responses replace this with the `requestedByUser` object (name + email).
+        `GET /v1/eventreports/{_id}/signedURL`; generation can take several minutes. The create
+        response includes `_requestedByUserId`; subsequent GET responses replace this with the
+        `requestedByUser` object (name + email).
       requestBody:
         required: true
         content:
@@ -94,7 +95,42 @@ paths:
         '401':
           $ref: '#/components/responses/401-unauthorized'
         '422':
-          $ref: '#/components/responses/422-unprocessable-entity'
+          description: |-
+            The window is out of bounds -- `startTime` more than 30 days back, `endTime` before
+            `startTime`, or a span over 3 days -- or `_flowIds` is empty. No report is created.
+          content:
+            application/json:
+              schema:
+                $ref: '#/components/schemas/Error'
+              examples:
+                span_over_three_days:
+                  summary: Window longer than 3 days
+                  value:
+                    errors:
+                      - field: endTime
+                        code: invalid_range
+                        message: For event report, the time gap between start and end time can't exceed 3 days.
+                start_over_thirty_days:
+                  summary: Start more than 30 days back
+                  value:
+                    errors:
+                      - field: startTime
+                        code: invalid_starttime
+                        message: The start time for the event report can't go beyond 30 days.
+                end_before_start:
+                  summary: endTime before startTime
+                  value:
+                    errors:
+                      - field: startTime
+                        code: invalid_range
+                        message: The end time can't be before than start time.
+                no_flows:
+                  summary: Empty _flowIds
+                  value:
+                    errors:
+                      - field: type
+                        code: invalid_flowevents
+                        message: Please select the flows for which the flow event report needs to be generated.
     get:
       x-internal: false
       operationId: listEventReports
@@ -227,6 +263,10 @@ paths:
         The URL is valid for approximately 15 minutes. Only works when the report `status` is
         `completed` -- requesting it in any other status returns 400. If the URL has expired,
         call this endpoint again to get a fresh one.
+
+        The file is served with `Content-Encoding: gzip`, so download it with a client that decodes
+        it (`curl --compressed`) or gunzip it afterwards. Its columns start `traceKey`, `time`,
+        `type`, `timeTaken`, `_flowJobId`.
       parameters:
         - name: _id
           in: path
@@ -670,19 +710,6 @@ components:
                 errors:
                   - code: invalid_request
                     message: The request body failed validation.
-    422-unprocessable-entity:
-      description: |
-        Unprocessable entity. The request was well-formed but was unable to be followed due to semantic errors.
-      content:
-        application/json:
-          schema:
-            $ref: '#/components/schemas/Error'
-          examples:
-            default:
-              value:
-                errors:
-                  - code: unprocessable_entity
-                    message: The request could not be processed due to a semantic error.
     404-not-found:
       description: |
         Not found. The requested resource does not exist or is not visible to the caller.

package/dist/export.yml

@@ -5580,7 +5580,10 @@ components:
               description: |-
                 When true, a file whose records partially fail parsing against the file definition
                 still emits the records that did parse, instead of failing the entire file. Applies to
-                file-definition (fixed-width, EDI) parsing only.
+                file-definition (fixed-width, EDI) parsing only. For X12 and EDIFACT the unit is the
+                transaction set (ST–SE, UNH–UNT), and it takes effect only with an EDI profile linked:
+                a failed set is set aside and reported as an error while the file's other sets go on
+                as records. Without this flag, one failed set fails the whole file.
               examples:
                 - true
                 - false

package/dist/filedefinition.yml

@@ -1163,8 +1163,12 @@ components:
             Primarily relevant for EDI EDIFACT format.
         skipEDIValidation:
           type: boolean
+          default: true
           description: |
-            Whether to skip EDI structural validation during parsing.
+            Whether to skip EDI element validation during parsing: element lengths, data types,
+            code lists, mandatory elements and element relationships. Omitted means skipped, so
+            these checks run only when it is `false`. Profile checks (`skipEDIProfileValidation`)
+            and missing mandatory segments are separate and still apply.
         skipEDIProfileValidation:
           type: boolean
           default: false
@@ -1188,6 +1192,11 @@ components:
           `maxOccurrence` is 1 or unset, matches only the first row: in fixed-width and delimited
           files the parser drops the rest without an error, and an X12 parse fails at the next row,
           naming a mandatory segment as missing.
+        - |-
+          Cloning an export that uses a file definition also creates a new file definition, with the
+          same name, and points the cloned export at it: the export's clone preview lists it as a
+          `FileDefinition` object. Two definitions with one name can follow, so tell them apart by
+          `_id`.
     ResourceResponse:
       type: object
       description: Response

package/dist/integration.yml

@@ -1024,97 +1024,6 @@ paths:
                     errors:
                       - code: invalid_ref
                         message: Integration not found.
-  /v1/integrations/{_id}/clonefamily:
-    get:
-      x-internal: false
-      operationId: listIntegrationCloneFamily
-      tags:
-        - Integrations
-      summary: List an integration's clone family
-      description: |-
-        Returns every member of the integration's clone family -- the original plus every clone
-        (recursively, across environments). Used by ILM to decide which integrations can be pull/push
-        sources or targets for the current integration.
-
-        This endpoint requires the `integrator-ashareid` header to establish a shared-account context.
-        For simpler clone-relationship lookups, the integration's own `_sourceId` field points to the
-        parent it was cloned from, and `GET /v1/integrations?_sourceId=<id>` lists the children.
-      parameters:
-        - name: integrator-ashareid
-          in: header
-          required: true
-          description: |-
-            Account-share id that scopes the request to a specific shared-account context. The bearer
-            token's user must be the `sharedWithUser` of this ashare.
-          schema:
-            type: string
-            format: objectId
-            examples:
-              - 6798f4dc43db9184f4bc8ee0
-        - name: _id
-          in: path
-          required: true
-          description: Integration id.
-          schema:
-            type: string
-            format: objectId
-            x-celigo-refModel: integrations
-            examples:
-              - 6842261335b64c0bcb308e4f
-      responses:
-        '200':
-          description: Clone family members.
-          content:
-            application/json:
-              schema:
-                $ref: '#/components/schemas/CloneFamilyResponse'
-              examples:
-                family:
-                  summary: Original integration plus one clone in another environment
-                  value:
-                    - _id: 6842261335b64c0bcb308e4f
-                      name: Shopify → NetSuite Orders
-                    - _id: 68f52162f00a35a1b26eaafb
-                      name: Clone - Shopify → NetSuite Orders
-                      _envId: 67b123456789abcdef012345
-        '401':
-          $ref: '#/components/responses/401-unauthorized'
-        '403':
-          description: |-
-            Either the `integrator-ashareid` header is malformed (code: `invalid_header_value`) or it
-            was omitted entirely (code: `get_not_allowed`, "GET resource not allowed for the given
-            model.").
-          content:
-            application/json:
-              schema:
-                $ref: '#/components/schemas/Error'
-              examples:
-                get_not_allowed:
-                  summary: Header omitted entirely
-                  value:
-                    errors:
-                      - code: get_not_allowed
-                        message: GET resource not allowed for the given model.
-                invalid_header_value:
-                  summary: Header malformed (not a 24-char hex id)
-                  value:
-                    errors:
-                      - code: invalid_header_value
-                        message: The value provided for request header Integrator-AShareId is invalid.
-        '404':
-          description: |-
-            Integration unknown, or the `integrator-ashareid` references a share whose `sharedWithUser`
-            isn't the bearer token's user (code: `access_restricted`).
-          content:
-            application/json:
-              schema:
-                $ref: '#/components/schemas/Error'
-              examples:
-                access_restricted:
-                  value:
-                    errors:
-                      - code: access_restricted
-                        message: Your access to an integrator.io environment or account has been disabled. For more information, contact account administrator.
   /v1/integrations/{_id}/flows:
     get:
       x-internal: false
@@ -2575,103 +2484,6 @@ paths:
           $ref: '#/components/responses/401-unauthorized'
         '404':
           $ref: '#/components/responses/404-not-found'
-  /v1/integrations/{_id}/iLMIgnore:
-    get:
-      x-internal: false
-      operationId: getIntegrationILMIgnoreFields
-      tags:
-        - Integrations
-      summary: List available ILM ignore fields
-      description: |-
-        Returns the catalog of *available* fields that can be ignored when pulling changes between
-        related integrations. Keys are lowercase resource types (`integration`, `flow`, `export`,
-        `import`, `connection`, `script`, ...); values are arrays of dot-path field names that can be
-        excluded from an ILM pull/revert. The key set is platform-driven — treat the response, not
-        the example list, as the authoritative catalog.
-
-        This endpoint requires the `integrator-ashareid` header to establish a shared-account context.
-        The current ignore configuration lives on the integration's `iLMIgnore` field (readable via
-        `GET /v1/integrations/{_id}`); this endpoint only returns the menu of what *can* be ignored.
-        To change the configuration, PATCH the integration with
-        `[{op:"replace", path:"/iLMIgnore", value:{...}}]`.
-      parameters:
-        - name: integrator-ashareid
-          in: header
-          required: true
-          description: |-
-            Account-share id that scopes the request to a specific shared-account context. The bearer
-            token's user must be the `sharedWithUser` of this ashare.
-          schema:
-            type: string
-            format: objectId
-            examples:
-              - 6798f4dc43db9184f4bc8ee0
-        - name: _id
-          in: path
-          required: true
-          description: Integration id.
-          schema:
-            type: string
-            format: objectId
-            x-celigo-refModel: integrations
-            examples:
-              - 6842261335b64c0bcb308e4f
-      responses:
-        '200':
-          description: Available ignore fields per resource type.
-          content:
-            application/json:
-              schema:
-                $ref: '#/components/schemas/ILMIgnoreResponse'
-              examples:
-                default:
-                  value:
-                    integration:
-                      - name
-                      - description
-                    flow:
-                      - disabled
-                      - name
-                    export:
-                      - name
-                      - http.body
-        '401':
-          $ref: '#/components/responses/401-unauthorized'
-        '403':
-          description: |-
-            Either the `integrator-ashareid` header is malformed (code: `invalid_header_value`) or it
-            was omitted entirely (code: `get_not_allowed`).
-          content:
-            application/json:
-              schema:
-                $ref: '#/components/schemas/Error'
-              examples:
-                get_not_allowed:
-                  summary: Header omitted entirely
-                  value:
-                    errors:
-                      - code: get_not_allowed
-                        message: GET resource not allowed for the given model.
-                invalid_header_value:
-                  summary: Header malformed (not a 24-char hex id)
-                  value:
-                    errors:
-                      - code: invalid_header_value
-                        message: The value provided for request header Integrator-AShareId is invalid.
-        '404':
-          description: |-
-            Integration unknown, or the `integrator-ashareid` references a share whose `sharedWithUser`
-            isn't the bearer token's user (code: `access_restricted`).
-          content:
-            application/json:
-              schema:
-                $ref: '#/components/schemas/Error'
-              examples:
-                access_restricted:
-                  value:
-                    errors:
-                      - code: access_restricted
-                        message: Your access to an integrator.io environment or account has been disabled. For more information, contact account administrator.
   /v1/integrations/{_id}/template:
     get:
       x-internal: false
@@ -4957,36 +4769,6 @@ components:
             type: string
             examples:
               - '6044134138'
-    CloneFamilyResponse:
-      type: array
-      description: |-
-        Members of an integration's clone family — the original integration plus every clone (including
-        clones of clones, across environments). Used by ILM to decide which integrations can pull from or
-        push to the current one.
-      items:
-        type: object
-        description: One integration in the clone family.
-        properties:
-          _id:
-            type: string
-            format: objectId
-            description: Integration id.
-            x-celigo-refModel: integrations
-            examples:
-              - 6842261335b64c0bcb308e4f
-          name:
-            type: string
-            description: Display name of the integration.
-            examples:
-              - Shopify → NetSuite Orders
-          _envId:
-            type: string
-            format: objectId
-            x-celigo-refModel: environments
-            examples:
-              - 67b123456789abcdef012345
-            description: Environment id the integration belongs to.
-        additionalProperties: true
     LookupCacheBase:
       type: object
       description: Writable fields shared by the request and response schemas.
@@ -5345,38 +5127,6 @@ components:
             - To register an existing topic instead, use `PUT /v1/integrations/{_id}/topics/{_topicId}/register`.
           examples:
             - - 6ab1f4a138d7733b45998d99
-    ILMIgnoreResponse:
-      type: object
-      description: |-
-        Catalog of resource-type → field-path lists that are *available* to be ignored during ILM pulls
-        and reverts. This is the full menu of what can be configured — to see the current ignore
-        configuration, read the `iLMIgnore` field on the integration itself via
-        `GET /v1/integrations/{_id}`.
-
-        Keys are lowercase resource type names (e.g. `integration`, `flow`, `export`, `import`,
-        `connection`, `script`). Each value is an array of dot-path strings naming fields that can be
-        excluded. Only resource types that have configurable ignores appear as keys.
-      additionalProperties:
-        type: array
-        description: Field paths (dotted) that can be ignored for this resource type.
-        items:
-          type: string
-      example:
-        integration:
-          - settings.general.description
-          - settings.general.name
-        flow:
-          - scheduling.startTime
-          - disabled
-        export:
-          - http.relativeURI
-          - delta.dateField
-        import:
-          - http.relativeURI
-          - mapping
-        connection:
-          - http.baseURI
-          - http.auth.token.token
     DependencyEntry:
       type: object
       description: A single resource that depends on the queried resource.

package/dist/mcp-server.yml

@@ -4679,6 +4679,14 @@ components:
             Do NOT include the protocol prefix (e.g., use "ftp.example.com", not "sftp://ftp.example.com").
           examples:
             - sftp.example.com
+          x-celigo-ai-guidance:
+            - |-
+              Don't point a connection at this environment's own Celigo file server (its
+              `endpoint.hostname`, with a username ending in `@<identifier>`, the server's own
+              `identifier`): a flow reads those files from Celigo Storage directly, and the save
+              refuses that pairing with 422 `fileserver_loopback_not_supported`. Another
+              environment's file server on the same host, such as a trading partner's, is an
+              ordinary connection.
         port:
           type: number
           description: |
@@ -4802,7 +4810,8 @@ components:
           description: |
             Maximum number of concurrent file transfer operations.
             FTP servers often have low connection limits — keep this value conservative.
-            Send no more than the account's licensed maximum (25 standard, 50 with an Environments license).
+            Send no more than the account's licensed maximum, the license's `concurrency` (25 standard,
+            50 with an Environments license).
             Unlike the rest of the `ftp` object, a `PUT` that omits this field keeps the stored value instead of clearing it.
           minimum: 1
           maximum: 50
@@ -4826,7 +4835,7 @@ components:
           description: |-
             Number of parallel transfer threads for this FTP connection. Must be a
             whole number — decimals are rejected on save. Send no more than the
-            account's licensed concurrency maximum. On
+            account's licensed concurrency maximum, the same as for `concurrencyLevel`. On
             save the value is mirrored into `userConcurrencyLevel` (and into
             `userTargetConcurrencyLevel` when auto-recovery of rate-limit errors
             is active with `targetConcurrencyLevel` set); clearing it clears the

package/dist/recyclebin.yml

@@ -419,9 +419,13 @@ paths:
         Resources are auto-purged after 30 days. A purge removes only the
         named resource: purge the bin entries that still reference it first
         (a deleted flow references its integration and the exports and imports
-        it used). An integration's Celigo Storage system folder is the
-        exception: the folder and everything in it go to the bin with the
-        integration and are purged with it.
+        it used). Purging a flow leaves its integration, exports, imports and
+        scripts where they are. A few purges remove more than the named
+        document: an integration's Celigo Storage system folder (with
+        everything in it) goes to the bin with the integration and is purged
+        with it; purging a connection first deletes the end users' own copies
+        of it (their MCP connections) for good; and purging a topic deletes
+        its stored messages.
       parameters:
         - $ref: '#/components/parameters/ResourceType'
         - name: _id

package/dist/state.yml

@@ -35,6 +35,10 @@ tags:
       Values must be non-empty JSON objects or arrays. Bare primitives,
       `null`, and empty containers (`{}`, `[]`) are rejected.
 
+      Each export, import or integration holds at most 10 keys, and the
+      global scope at most 100. A new key past the limit is refused with 422
+      `limit_reached`; replacing an existing key still works.
+
       Prefer resource-scoped state over global state when the data belongs
       to a specific export, import, or integration. State keys are upserted —
       PUT creates if absent, replaces if present. There is no PATCH; send the
@@ -243,15 +247,25 @@ paths:
         '422':
           description: |-
             The request body is empty, or is an empty container (`{}` or
-            `[]`).
+            `[]`), or the key is new and the account already holds 100 global
+            keys.
           content:
             application/json:
               schema:
                 $ref: '#/components/schemas/Error'
-              example:
-                errors:
-                  - code: missing_request_body
-                    message: Please send state as request body in the PUT request.
+              examples:
+                missing_request_body:
+                  summary: Empty body or empty container
+                  value:
+                    errors:
+                      - code: missing_request_body
+                        message: Please send state as request body in the PUT request.
+                limit_reached:
+                  summary: The account already holds 100 global keys
+                  value:
+                    errors:
+                      - code: limit_reached
+                        message: 'Cannot save state, allowed number of states already reached for user : 60f1a2b3c4d5e6f7a8b9c0d1'
     delete:
       x-internal: false
       operationId: deleteGlobalStateKey
@@ -710,15 +724,24 @@ paths:
         '422':
           description: |-
             The request body is empty, or is an empty container (`{}` or
-            `[]`).
+            `[]`), or the key is new and the resource already holds 10 keys.
           content:
             application/json:
               schema:
                 $ref: '#/components/schemas/Error'
-              example:
-                errors:
-                  - code: missing_request_body
-                    message: Please send state as request body in the PUT request.
+              examples:
+                missing_request_body:
+                  summary: Empty body or empty container
+                  value:
+                    errors:
+                      - code: missing_request_body
+                        message: Please send state as request body in the PUT request.
+                limit_reached:
+                  summary: The resource already holds 10 keys
+                  value:
+                    errors:
+                      - code: limit_reached
+                        message: 'Cannot save state, allowed number of states already reached for export : 66a1f2c3b4d5e6f7a8b9c0d1'
     delete:
       x-internal: false
       operationId: deleteResourceStateKey

package/dist/sync.yml

@@ -242,7 +242,10 @@ paths:
           description: |-
             Validation failed. `source._connectionId` is missing
             (`missing_source_connection`), the integration was not created with
-            `syncs: true` (`integration_not_for_sync`), the destination duplicates another sync
+            `syncs: true` (`integration_not_for_sync`), `destination._connectionId` isn't a
+            connection in the account (`missing_destination_connection`), the destination
+            connection is offline while a `database` or `schema` is set
+            (`offline_destination_connection`), the destination duplicates another sync
             (`duplicate_destination_host`), the destination connector rejected the
             sync configuration (`invalid_destination_sync_config`), two enabled
             datasets resolve to the same destination table
@@ -444,14 +447,21 @@ paths:
           description: |-
             Validation failed. `source._connectionId` is missing
             (`missing_source_connection`), the sync is not ready to enable
-            (`sync_not_enabled`), the destination duplicates another sync
+            (`sync_not_enabled`), the update drops the saved destination connection or names a
+            connection that doesn't exist (`missing_destination_connection`, "Destination
+            connection is missing in the request"), the new destination connection is another
+            warehouse type than the saved one (`duplicate_destination_type`, "Destination
+            connection type is not same as the previous destination connection type"), a new
+            `database` or `schema` was sent while the destination connection is offline
+            (`offline_destination_connection` — the platform creates the schema when the
+            destination is saved), the destination duplicates another sync
             (`duplicate_destination_host`), the destination connector rejected the
             sync configuration (`invalid_destination_sync_config`), two enabled
             datasets resolve to the same destination table
             (`conflicting_table_name`), a field exceeds its size limit
             (`size_limit_exceeded`), the drift policy is inconsistent
             (`drift_policy_invalid_setup`), or the destination's application differs from
-            the integration's other syncs or from the sync's saved destination.
+            the integration's other syncs.
           content:
             application/json:
               schema:
@@ -798,8 +808,9 @@ paths:
           description: |-
             Bad request. The body is not a non-empty array (`INVALID_REQUEST_BODY` /
             `EMPTY_REQUEST_BODY`), an item has both or neither of `_id`/`externalId`
-            (`dataset_externalId_id_required`), an `externalId` already exists or an `_id`
-            does not belong to the sync, `tableName` values collide
+            (`dataset_externalId_id_required`), an `externalId` already exists on the sync
+            (`create_dataset_present`) or an `_id` does not belong to it
+            (`update_dataset_not_present`), `tableName` values collide
             (`duplicate_dataset_table_names`), or an export dataset's export uses a
             different connection than the sync (`connection_mismatch`).
           content:
@@ -3016,8 +3027,10 @@ paths:
                       deltaFields:
                         type: array
                         description: |-
-                          Column names usable as the incremental-extraction cursor
-                          (`exportProperties.delta.dateField`).
+                          Column names usable as the incremental-extraction cursor. A
+                          Salesforce dataset uses the first entry itself; a NetSuite
+                          saved-search dataset names its cursor in
+                          `exportProperties.delta.dateField`.
                         items:
                           type: string
                         examples:
@@ -3438,7 +3451,8 @@ paths:
                           type: boolean
                           description: |-
                             Only present (as `true`) when the export no longer uses the
-                            sync's source connection.
+                            sync's source connection, or is no longer among the source
+                            connection's exports (for example, it was deleted).
                           examples:
                             - true
                         totalFieldCount:
@@ -4421,8 +4435,10 @@ components:
                   type: string
                   description: |-
                     Source field that carries each record's last-modified timestamp, used as
-                    the delta cursor. Required for NetSuite datasets that extract via a
-                    `netsuite.recordType`; Salesforce datasets derive it automatically.
+                    the delta cursor. Used by NetSuite saved-search datasets, where it is
+                    required when a `netsuite.recordType` is set. Salesforce datasets take the
+                    first entry of the column read's `deltaFields` instead, and HTTP-connector
+                    datasets have no date field: their delta filter is part of the request.
                   examples:
                     - lastmodifieddate
             netsuite:
@@ -4431,9 +4447,13 @@ components:
               properties:
                 recordType:
                   type: string
+                  writeOnly: true
                   description: |-
                     NetSuite record type backing a saved-search dataset. When set with
-                    `exportProperties.type: delta`, a `delta.dateField` is required.
+                    `exportProperties.type: delta`, a `delta.dateField` is required. Write-only:
+                    dataset reads don't return it, so an update item built from a read-back
+                    clears it; the sync's next run restores a missing value from the saved
+                    search's catalog entry (`appSpecificProperties.recordType`).
                   examples:
                     - salesorder
         driftPolicy:
@@ -4466,8 +4486,10 @@ components:
           readOnly: true
           description: |-
             Only present (as `true`) in list responses, on export-backed datasets whose
-            export no longer uses the sync's source connection. Repoint the export or
-            recreate the dataset to resume replicating it.
+            export no longer uses the sync's source connection. The source catalog listing
+            also flags an export-backed dataset whose export it no longer lists, such as a
+            deleted export. Repoint the export or recreate the dataset to resume replicating
+            it.
           examples:
             - true
     DatasetUpsertItem:
@@ -4595,8 +4617,10 @@ components:
                   type: string
                   description: |-
                     Source field carrying each record's last-modified timestamp, used as the
-                    delta cursor. Required for NetSuite datasets extracted via a
-                    `netsuite.recordType`; Salesforce datasets derive it automatically.
+                    delta cursor. Used by NetSuite saved-search datasets, where it is required
+                    when a `netsuite.recordType` is set; it isn't checked against the column
+                    read's `deltaFields`. Salesforce datasets take the first `deltaFields` entry
+                    instead, and HTTP-connector datasets take no date field.
                   examples:
                     - lastmodifieddate
             netsuite:
@@ -4606,8 +4630,11 @@ components:
                 recordType:
                   type: string
                   description: |-
-                    NetSuite record type backing a saved-search dataset. When set with
-                    `exportProperties.type: delta`, a `delta.dateField` is required.
+                    NetSuite record type backing a saved-search dataset, from the saved search's
+                    catalog entry (`appSpecificProperties.recordType`). When set with
+                    `exportProperties.type: delta`, a `delta.dateField` is required. Reads never
+                    return it, so an update item built from a read-back clears it unless you add
+                    it back; the sync's next run restores a missing value from the catalog.
                   examples:
                     - salesorder
         driftPolicy:
@@ -4649,7 +4676,11 @@ components:
             Scopes this run to these datasets on the sync. Omit the field to run every
             enabled dataset; an empty array is rejected. Each id must belong to this sync
             and be enabled, and subset selection on a normal run is rejected unless the
-            account has partial normal sync enabled.
+            account has partial normal sync enabled. A selection larger than the per-run
+            maximum is accepted here but stops the run as it starts: no table runs, and
+            the sync's errors record `max_datasets_exceeded` ("Selected datasets exceed the
+            maximum allowed per run (N). Reduce the selection and retry."). Split a large
+            selection across runs.
           items:
             type: string
             format: objectId

package/dist/tool.yml

@@ -8187,6 +8187,14 @@ components:
             Do NOT include the protocol prefix (e.g., use "ftp.example.com", not "sftp://ftp.example.com").
           examples:
             - sftp.example.com
+          x-celigo-ai-guidance:
+            - |-
+              Don't point a connection at this environment's own Celigo file server (its
+              `endpoint.hostname`, with a username ending in `@<identifier>`, the server's own
+              `identifier`): a flow reads those files from Celigo Storage directly, and the save
+              refuses that pairing with 422 `fileserver_loopback_not_supported`. Another
+              environment's file server on the same host, such as a trading partner's, is an
+              ordinary connection.
         port:
           type: number
           description: |
@@ -8310,7 +8318,8 @@ components:
           description: |
             Maximum number of concurrent file transfer operations.
             FTP servers often have low connection limits — keep this value conservative.
-            Send no more than the account's licensed maximum (25 standard, 50 with an Environments license).
+            Send no more than the account's licensed maximum, the license's `concurrency` (25 standard,
+            50 with an Environments license).
             Unlike the rest of the `ftp` object, a `PUT` that omits this field keeps the stored value instead of clearing it.
           minimum: 1
           maximum: 50
@@ -8334,7 +8343,7 @@ components:
           description: |-
             Number of parallel transfer threads for this FTP connection. Must be a
             whole number — decimals are rejected on save. Send no more than the
-            account's licensed concurrency maximum. On
+            account's licensed concurrency maximum, the same as for `concurrencyLevel`. On
             save the value is mirrored into `userConcurrencyLevel` (and into
             `userTargetConcurrencyLevel` when auto-recovery of rate-limit errors
             is active with `targetConcurrencyLevel` set); clearing it clears the
@@ -30253,7 +30262,10 @@ components:
               description: |-
                 When true, a file whose records partially fail parsing against the file definition
                 still emits the records that did parse, instead of failing the entire file. Applies to
-                file-definition (fixed-width, EDI) parsing only.
+                file-definition (fixed-width, EDI) parsing only. For X12 and EDIFACT the unit is the
+                transaction set (ST–SE, UNH–UNT), and it takes effect only with an EDI profile linked:
+                a failed set is set aside and reported as an error while the file's other sets go on
+                as records. Without this flag, one failed set fails the whole file.
               examples:
                 - true
                 - false

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@celigo/api-specs",
-  "version": "0.2.167",
+  "version": "0.2.169",
   "private": false,
   "description": "Public OpenAPI bundles (dist/) and the celigo-cli lint schema pack (schemas.json).",
   "keywords": [

package/schemas.json

@@ -1,7 +1,7 @@
 {
 	"provenance": {
 		"repo": "celigo/integrator-api-specs",
-		"iasCommit": "560182fc",
+		"iasCommit": "750b2035",
 		"generatedAt": "2026-10-22"
 	},
 	"roots": {