Experience · Software component
Agent User Interface
Software componentExperienceExperience & Human OversightVariation point (abstract)arc:AgentUserInterface
An abstract user-facing interaction surface through which humans direct agents and perceive agent status, reasoning, uncertainty, evidence, and results.
Responsibility. Mediates human-agent interaction for a chosen interaction model.
Also known as: Agent UI, Agent interface
Variants
| Variant | When to choose |
|---|---|
| Command Palette with Agent Suggestions | Choose for IDEs, productivity and design tools where power users value keyboard speed and the agent can predict intent from context; avoid when discoverability for new users, multi-turn dialogue, or extensive explanation is required. |
| Conversational (Chat) Interface | Choose for general Q&A, customer support, simple task automation, and exploratory conversations; avoid for workflows needing visual process representation, dense structured data, or parallel information streams. |
Relationships
is configured by structural
invokes dependency
is invoked by dependency
writes dependency
receives data from dynamic
is audited by assurance
Design guidance
- SHOULD show final results by default and reveal reasoning and technical details on demand (progressive disclosure).
- SHOULD always display current agent status during processing.
- SHOULD calibrate explanation depth to decision impact rather than treating all decisions equally.
- MUST be fully operable by keyboard and compatible with assistive technologies via semantic markup and live regions.
- MUST NOT convey status, confidence, or risk by color alone.
Quantitative guidance
As stated by the sources; verify before use.
- Working memory holds approximately seven (plus or minus two) items; displaying every intermediate step violates this limit (Ch1.1A).
- WCAG requires a minimum contrast ratio of 4.5:1 for normal text and 3:1 for large text (Ch1.1B).
Classification
- Patterns
- Progressive disclosureTransparency and explainabilityContext awareness and continuityStatus visibility
- Quality attributes
- Interaction capability (ISO/IEC 25010)Transparency and accountability (NIST AI RMF: accountable and transparent)
- Risks mitigated
- Cognitive overloadUser trust lossSilent system abandonment
- Frameworks & regulations
- WCAG 2.1 Level AA
Sources
- Ch1.1A: T. Nguyen, "Designing User Interfaces for Intuitive Human-Agent Interaction," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 1.1A. ISBN: 9798244538229.
- Ch1.1B: T. Nguyen, "Human-in-the-Loop Patterns and Accessible Design," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 1.1B. ISBN: 9798244538229.
- Ch9.4: T. Nguyen, "Fairness and Bias Mitigation," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 9.4. ISBN: 9798244538229.
- Ref9.02: "Responsible AI and Ethical Principles," unpublished reference note (02-Responsible-AI-Ethical-Principles.md), Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam supplementary materials, 2026. unpublished note