Intelligent Task Orchestration
Agent doesn't follow fixed scripts. It decomposes tasks on its own, runs multiple steps concurrently, and auto-adjusts plans when problems arise — like a thinking executor, not a button-pressing robot.
Standard Mode
availableAgent thinks and acts step by step. After each step, it decides what to do next based on results. Ideal for tasks requiring judgment and trial-and-error, like alert investigation, document analysis, and information retrieval.
Planning Mode
availableAgent first creates a complete execution plan, decomposing tasks into steps and identifying dependencies — independent steps run concurrently, dependent steps run sequentially. After completion, it checks results and auto-retries with adjusted plans if goals aren't met.
| Standard Mode | Planning Mode | |
|---|---|---|
| How it works | Step-by-step thinking and acting | Plan first, then parallel execution |
| Best for | Tasks requiring judgment and trial-and-error | Multi-step, parallelizable tasks |
| Self-correction | Auto-recovers from errors | Auto re-plans when goals aren't met |
| Visualization | Step-by-step reasoning display | Interactive flowchart with real-time status |
Visual Pipeline (Blueprint Mode)
comingFor teams who prefer drag-and-drop, a visual pipeline editor with three progressive levels:
Built-in scenario templates (contract review, financial reconciliation, approval assistance, etc.), forkable and customizable.
Context Management
availableLong conversations auto-compress earlier content while preserving recent text verbatim. Critical information (task descriptions, user corrections) is protected from compression. A unified token budget manager ensures conversations always stay within the model's context window.