Home > Tasks > Reasoning > TableCompilationTask

TableCompilationTask

Generate structured tables with AI-computed cell values using parallelized partition processing and multi-format output support.

Category: Reasoning Parallel Processing Output: HTML/CSV/JSON

⚙️ ExecutionConfig.json

{
  "rows": ["Apple", "Tesla", "Microsoft"],
  "columns": ["Industry", "Primary Product"],
  "cell_query": "What is the {column} for {row}?",
  "task_description": "Tech giants comparison"
}

→

👁️ Session UI (Table Tab)

	Industry	Primary Product
Apple	Consumer Electronics	iPhone
Tesla	Automotive / Energy	Electric Vehicles
Microsoft	Software / Cloud	Windows & Azure

Test Workspace Artifacts

Execution Configuration

Field	Type	Description
`rows` *	`List<String>`	Row headers for the table.
`columns` *	`List<String>`	Column headers for the table.
`cell_query` *	`String`	Template for cell generation. Use `{row}` and `{column}` placeholders.
`task_description`	`String`	Overall context or description for the table generation.

Field	Default	Description
`partition_size`	`2`	Maximum partition size for parallel processing (e.g., 2 means 2x2 partitions). Range: 1-10.

Embedded Execution (Headless)

Use the UnifiedHarness to run this task programmatically in CI/CD or CLI tools.

import com.simiacryptus.cognotik.plan.TaskType
import com.simiacryptus.cognotik.plan.tools.reasoning.TableCompilationTask.Companion.TableCompilation
import com.simiacryptus.cognotik.plan.tools.reasoning.TableCompilationTask.TableCompilationTaskExecutionConfigData

fun generateComparisonTable(harness: UnifiedHarness, projectDir: File) {
    val config = TableCompilationTaskExecutionConfigData(
        rows = listOf("Performance", "Security", "Cost"),
        columns = listOf("Solution A", "Solution B"),
        cell_query = "Compare {column} against {row} criteria.",
        task_description = "Vendor Analysis"
    )

    harness.runTask(
        taskType = TableCompilation,
        executionConfig = config,
        workspace = projectDir,
        autoFix = true
    )
}

Prompt Segment

The following logic is injected into the orchestrator's context:

TableCompilation - Generate structured tables with AI-computed cell values
  ** Specify row headers in the 'rows' array
  ** Specify column headers in the 'columns' array
  ** Provide a cell_query template using {row} and {column} placeholders
  ** Cells are computed in partitions for efficiency (configurable partition_size)
  ** Supports output formats: markdown, html, csv

TableCompilationTask

Test Workspace Artifacts

Execution Configuration

Type Configuration

Lifecycle

1. Partitioning

2. Batch Prompting

3. Parallel Execution

4. Multi-Format Rendering

Embedded Execution (Headless)

Prompt Segment