update di plugin reference

Author: seebi

data/plugins.json

@@ -47,6 +47,7 @@ nav:
     - "Send email": SendEMail.md
     - "Send Mattermost messages": cmem_plugin_mattermost.md
     - "Set or Overwrite parameters": cmem_plugin_parameters-ParametersPlugin.md
+    - "Set parameters": setParameters.md
     - "SHACL validation with pySHACL": shacl-pyshacl.md
     - "SOQL query (Salesforce)": cmem_plugin_salesforce-SoqlQuery.md
     - "Spark SQL query": CustomSQLExecution.md
@@ -58,7 +59,7 @@ nav:
     - "Start Workflow per Entity": cmem_plugin_loopwf-task-StartWorkflow.md
     - "Store Vector Embeddings": cmem_plugin_pgvector-Store.md
     - "Unpivot": Unpivot.md
-    - "Update SemSpect": cmem_plugin_semspect-task-Update.md
+    - "Update Graph Insights Snapshots": cmem_plugin_graph_insights-Update.md
     - "Upload File to Knowledge Graph": eccencaDataPlatformGraphStoreFileUploadOperator.md
     - "Upload files to Nextcloud": cmem_plugin_nextcloud-Upload.md
     - "Upload local files": cmem_plugin_project_resources-UploadLocalFiles.md

docs/build/reference/customtask/.pages

@@ -0,0 +1,65 @@
+---
+title: "Update Graph Insights Snapshots"
+description: "Update one or more snapshots, optionally selected by affected graph."
+icon: octicons/cross-reference-24
+tags: 
+    - WorkflowTask
+    - PythonPlugin
+---
+# Update Graph Insights Snapshots
+<!-- This file was generated - DO NOT CHANGE IT MANUALLY -->
+
+!!! note inline end "Python Plugin"
+
+    This operator is part of a Python Plugin Package.
+    In order to use it, you need to install it,
+    e.g. with cmemc.
+
+This workflow task updates [Graph Insights](https://go.eccenca.com/feature/explore-graph-exploration-graph-insights?lang=en&origin=cmem-plugin-graph-insights)
+    snapshots for a specified graph in your system.
+
+## Behavior
+
+- **No graph selected**: All snapshots in the system are updated
+- **Graph selected**: Only snapshots associated with the selected graph are updated
+
+## Usage
+
+1. Add this task to your workflow.
+2. Optionally select a specific graph to limit which snapshots are updated.
+3. Use the "Preview Snapshots" action to see which snapshots will be affected before execution.
+4. Run the workflow to update the snapshots.
+
+## Prerequisites
+
+- Graph Insights must be active in your system
+- User must have permissions to access Graph Insights
+- The plugin will skip execution with a warning if these conditions are not met
+    
+
+## Parameter
+
+### Selected Graph
+
+Selected graph to update snapshots for. Leave empty for updating all snapshots.
+
+- ID: `selected_graph`
+- Datatype: `string`
+- Default Value: `None`
+
+
+
+
+
+## Advanced Parameter
+
+### Timeout
+
+Timeout in seconds for Graph Insights API.
+
+- ID: `timeout`
+- Datatype: `double`
+- Default Value: `100`
+
+
+

docs/build/reference/customtask/cmem_plugin_graph_insights-Update.md

@@ -35,32 +35,30 @@ AI-powered text generation, analysis, and transformation tasks within Corporate
 ## Input/Output Behavior
 
 After processing, each entity receives an additional path (default: `_instruction_output`)
-containing the LLM response. Input/output ports are automatically configured based on
-template variables:
+containing the LLM response. For TEXT and JSON_MODE output formats, the response is stored
+in this path. For STRUCTURED_OUTPUT, the Pydantic model fields are directly added to the
+entity (the `_instruction_output` path is not used).
+
+Input/output ports are automatically configured based on template variables:
 
 - **No placeholders**: No input ports required
-- **With placeholders**: Dynamic input ports created for each template variable
-- **Port ordering**: Variables sorted alphabetically determine port order
+- **With placeholders**: Single input port created for entity data
 - **Schema handling**: Fixed schemas when using specific entity paths, flexible schemas otherwise
 
 ## Template System
 
 Uses Jinja2 templating for dynamic prompts:
 
 ```jinja2
-{{ variable }}           # Entire entity as JSON
-{{ variable.name }}      # Specific entity property
-{{ variable_a.title }}   # Property from first additional input port
-{{ variable_b.content }} # Property from second additional input port
+{{ entity }}           # Entire entity as JSON
+{{ entity.name }}      # Specific entity property
 ```
 
-The followin template processing rule are implemented:
+The following template processing rules are implemented:
 
 1. **Variable Extraction**: Automatically detects template variables to configure input ports
-2. **Entity Iteration**: Main processing iterates over first input port entities
-3. **Additional Inputs**: Secondary ports provide context data for template rendering
-4. **Consumption Modes**: Choose between first-entity or all-entities consumption from
-   additional ports
+2. **Entity Iteration**: Processes entities from the single input port individually
+3. **Single Entity Context**: Each entity is processed independently with its own template context
 
 ## Output Formats
 
@@ -167,7 +165,7 @@ An optional API key for authentication.
 
 ### Instruct Model
 
-The identifier of the instruct model to use. Note that some provider do not support a model list endpoint. Just create a custom entry then. Available model IDs for some public providers can be found here: [Claude](https://docs.claude.com/en/docs/about-claude/models/overview), [OpenRouter](https://openrouter.ai/models), [Azure](https://learn.microsoft.com/en-us/azure/ai-foundry/foundry-models/concepts/models-sold-directly-by-azure).
+The identifier of the instruct model to use. Note that some provider do not support a model list endpoint. Just create a custom entry then. Available model IDs for some public providers can be found here: [OpenAI](https://platform.openai.com/docs/models), [Claude](https://docs.claude.com/en/docs/about-claude/models/overview), [OpenRouter](https://openrouter.ai/models), [Azure](https://learn.microsoft.com/en-us/azure/ai-foundry/foundry-models/concepts/models-sold-directly-by-azure). **Note:** For STRUCTURED_OUTPUT format, only certain models support structured outputs. See [OpenAI Structured Outputs Guide](https://platform.openai.com/docs/guides/structured-outputs) for supported models.
 
 - ID: `model`
 - Datatype: `string`
@@ -221,7 +219,7 @@ The timeout for a single API request in seconds.
 
 ### Instruction Output Path
 
-The entity path where the instruction result will be provided.
+The entity path where the instruction result will be provided. Note: This parameter is not used when Output Format is set to STRUCTURED_OUTPUT. For structured outputs, only the Pydantic model fields are included in the output schema.
 
 - ID: `instruction_output_path`
 - Datatype: `string`
@@ -251,16 +249,6 @@ A list of messages comprising the conversation compatible with OpenAI chat compl
 
 
 
-### Consume all entities from additional input ports
-
-If enabled, all entities from additional input ports will be consumed. Otherwise, only the first entity of the additional ports will be used.
-
-- ID: `consume_all_entities`
-- Datatype: `boolean`
-- Default Value: `false`
-
-
-
 ### Output Format
 
 Specifying the format that the model must output. Possible values are `TEXT` - Standard text output, `STRUCTURED_OUTPUT` - output follows a given schema. Add your schema as Pydantic model in the parameter below, `JSON_MODE` - a more basic version of the structured outputs feature where you have to add your structure to the prompt template.

docs/build/reference/customtask/cmem_plugin_llm-ExecuteInstructions.md

@@ -1,6 +1,6 @@
 ---
 title: "SOQL query (Salesforce)"
-description: "Executes a custom Salesforce Object Query (SOQL) to return sets of data your organization’s Salesforce account."
+description: "Executes a custom Salesforce Object Query (SOQL) to return sets of data your organization's Salesforce account."
 icon: octicons/cross-reference-24
 tags: 
     - WorkflowTask
@@ -17,9 +17,9 @@ tags:
 
 
 This task executes a custom Salesforce Object Query (SOQL)
-and returns sets of tabular data from your organization’s Salesforce account.
+and returns sets of tabular data from your organization's Salesforce account.
 
-> Use the Salesforce Object Query Language (SOQL) to search your organization’s
+> Use the Salesforce Object Query Language (SOQL) to search your organization's
 > Salesforce data for specific information. SOQL is similar to the SELECT statement in
 > the widely used Structured Query Language (SQL) but is designed specifically for
 > Salesforce data.

docs/build/reference/customtask/cmem_plugin_salesforce-SoqlQuery.md

@@ -1,6 +1,6 @@
 ---
 title: "Create/Update Salesforce Objects"
-description: "Manipulate data in your organization’s Salesforce account."
+description: "Manipulate data in your organization's Salesforce account."
 icon: octicons/cross-reference-24
 tags: 
     - WorkflowTask
@@ -18,7 +18,7 @@ tags:
 
 This task retrieves data from an incoming workflow task (such as a SPARQL query),
 and sends bulk API requests to the Salesforce Object API, in order to
-manipulate data in your organization’s Salesforce account.
+manipulate data in your organization's Salesforce account.
 
 The working model is:
 - Each entity from the input data is interpreted as a single Salesforce object of the

docs/build/reference/customtask/cmem_plugin_salesforce-workflow-operations-SobjectCreate.md

@@ -25,28 +25,35 @@ JSON Schema specification.
 The used JSON Schema needs to be provided as a JSON Dataset in the project.
 
 ### Input Modes
+
 The plugin supports two input modes for validation:
+
 1. **Validate Entities**: Validates entities received from the input port in the workflow.
-2. **Validate JSON Dataset**: Validates a JSON dataset stored in the project. 
+2. **Validate JSON Dataset**: Validates a JSON dataset stored in the project.
    - If the JSON dataset is a JSON array, the schema will validate each object inside the array.
    - If the JSON dataset is a JSON object, it will be validated against the schema directly.
 
 Validated data objects can be sent to an output port for further processing in the workflow
 or saved in a JSON dataset in the project.
 
 ### Output Modes
-1. **Valid JSON objects sent to Output Port**: Valid JSON objects can be sent as entities to the output port.
-2. **Saved in JSON Dataset**: Valid JSON objects can be stored in a specified JSON dataset in the project.
+
+1. **Valid JSON objects sent to Output Port**: Valid JSON objects can be sent as entities
+   to the output port.
+2. **Saved in JSON Dataset**: Valid JSON objects can be stored in a specified JSON dataset
+   in the project.
 
 ### Error Handling
+
 The task can either:
+
 - Fail instantly if there is a data violation, halting the workflow.
-- Provide warnings in the workflow report, allowing follow-up tasks to run based on the validated data.
+- Provide warnings in the workflow report, allowing follow-up tasks to run based on the
+  validated data.
 
 The error handling behavior is configurable through the `Fail on violations` parameter.
 
 
-
 ## Parameter
 
 ### Source / Input Mode

docs/build/reference/customtask/cmem_plugin_semspect-task-Update.md

@@ -15,7 +15,24 @@ tags:
     In order to use it, you need to install it,
     e.g. with cmemc.
 
-This workflow task is basically a yaml2json command.
+
+This workflow task parses YAML content from multiple sources and converts it to various output
+formats.
+
+**Input Sources:**
+- **entities**: Parse YAML from input port entities in a workflow
+- **code**: Parse YAML from directly entered source code
+- **file**: Parse YAML from uploaded project file resources
+
+**Output Formats:**
+- **entities**: Convert parsed structure to entities for workflow processing
+- **json_entities**: Output as single JSON entity to the output port
+- **json_dataset**: Save parsed structure directly to a JSON dataset
+
+The plugin provides flexible YAML-to-JSON conversion with configurable input schema
+types and paths for entity-based processing. It includes comprehensive validation and
+error handling for all supported modes.
+
 
 ## Parameter