The Human-Agent Experience

We are moving beyond simple assistants. Today's agents act with greater autonomy, coordinate across systems, and collaborate with humans in nuanced ways. To support this shift, our product design team has developed research-backed principles and patterns for creating systems that are transparent, trustworthy, and truly collaborative.

Explore the SDK

Guiding principles

Five ideas for trustworthy agent experiences that prioritize human control and agency.


Control

Let the Human Set the Rules

?

What it means

Humans guide how agents operate by setting boundaries, preferences, and intent. Rather than making assumptions, agents respond to this direction with flexibility. Control doesn't mean limiting autonomy, it means aligning it with human goals.

!

Why this matters

People are more likely to trust and engage with agents when they understand how decisions are made, and can shape them. Giving users control over behaviors, permissions, and preferences makes the system feel collaborative and intentional, rather than unpredictable.

Interaction example: Instruction / Scope

1

Instruction mode

Users define interaction boundaries by selecting input modes, guiding the agent to operate safely within intended, user-controlled scopes.

2

Task specific boundaries

Specific checkboxes define what the AI is allowed to change (e.g., title edits) and what it must avoid. These help define clear behavioral constraints.

Instructions via
Click the tabs above to explore
1 of 3 Scope & Boundaries
Update the item label if it doesn't match the defined scope
Extend the description to clarify boundaries without altering original intent
2 of 3 Another Setting
3 of 3 Behavioral Limits
U
Turn off auto-updates for this app
Do you want to turn off auto-updates for all apps or just this one?
U
Yeah, show me the options
Which would you like?
Option 1 Option 2 Option 3

Overview

This configuration file defines preferences and settings for automated review and analysis of changes, with a focus on new settings. It supports automation, best practice recommendations, and change assessments.

Title and Description Guidelines

  • Title:
    • Update titles as needed to reflect the content of the changes.
  • Description:
    • Add relevant context or detail to the PR description directly; do not leave suggestions as comments.
    • Reference: How to write a good request description

Summary of Changes

1. What is the Instruction / Scope pattern in this context?

It's when the user gives a one-time instruction — such as changing a system setting — and the scope of the change is defined either in the file content or in the action of uploading the file itself.

2. What does a typical user action look like?

The user uploads a file (e.g., config.json, settings.yaml, or a .docx template) that contains configuration data or preferences

3. What does the system (agent) do upon receiving the file?

Validates the file (format, schema, safety)

Interaction example: Authority Sliders

Agent Control Panel

Click tabs or drag the slider to explore

Scope of Authority

Monitoring Guided Full Control
1

Monitoring

Agent observes and reports but takes no autonomous action. Full human oversight.

2

Guided

Agent suggests actions and awaits human approval before executing. Collaborative control.

3

Full Control

Agent acts autonomously within defined boundaries, optimizing for outcomes without step-by-step approval.

Interaction example: Kill switch and preview modes

1

Immediate agent shutdown

A prominent "Disable Agent" toggle gives users a fast, irreversible way to halt all agent activity. This supports emergency intervention and restores user authority instantly.

2

Visible control settings

The system shows settings upfront allowing the user to assess whether to make decisions based on risks and live situations.

Agent Control Panel

Disable Agent

Current Activities

Approval Load *

75%

How to implement

  • Make automation opt-in, never assume users want it
  • Let users override any system decision, anytime
  • Show what rules are active and how they're working
  • Make it easy to change and adjust settings

Common pitfalls

Over-reliance on implicit controls

Assuming users understand what the agent is doing without clear communication

Boundary creep

Agents gradually take on more than intended without safeguards or oversight

Ambiguous authority

It is unclear if the agent or the human is responsible for the task, especially in case of failure states

Hidden behavior

The agent performs actions without surfacing intent or outcomes to the user

Clarity

Show the agent's reasoning

?

What it means

Agents should make their reasoning, context, and confidence visible. Instead of acting like black boxes, they show how decisions are made so users can understand, question, or adjust them.

!

Why this matters

People make better decisions when they understand how an agent reached its conclusion. Clarity helps users spot errors, judge reliability, and decide when to lean in or push back.

Interaction example: Inline Rationale

1

Agent's reasoning surfaced

The agent labels its own decision logic, making invisible prioritization visible so users can understand, question, or reorder.

2

Scope of impact

Quantified impact badges let users grasp real-world consequences at a glance — without needing to investigate further.

3

Contextual rationale

A brief description explains the incident's cause, supporting informed decisions without overwhelming the interface.

Prioritizing incidents based on severity and impact
Incident 1
High Impact: 8 departments; 233 Devices

Description of the incident, what is causing it and more details on impact

Incident 2
Medium Impact: 3 departments; 75 Users

Description of the incident, what is causing it and more details on impact

Interaction example: Diagnostic Report

1

Reasoning linked to confidence

Explanations are paired with certainty indicators, showing why the system believes something and how strongly it believes it. This helps users validate or challenge the logic.

2

Confidence levels made visible

Each insight is accompanied by a clear degree of certainty, often a percentage or visual indicator. This helps users assess how much trust to place in each suggestion.

3

Actions calibrated to certainty

The system tailors its suggested actions based on confidence — more assertive steps when certainty is high, and more cautious ones when confidence is low.

4

Low confidence is still shown

Even uncertain insights are surfaced, not hidden — but clearly marked. This promotes transparency and allows human judgment to guide next steps.

Diagnostic Report with Actionables

Suspected Cause Confidence Rationale Recommended Action
High Resource Utilization in Component A
High (80%)
Reasoning on why this is happening
Environmental Interference in Zone X
Medium (43%)
Reasoning on why this is happening
Processing Bottleneck
Low (25%)
Reasoning on why this is happening

Interaction example: Findings with Sources

1

Source labels are visually distinct and clickable

The source elements are styled for immediate recognition and likely interactive (e.g., tags, badges, or buttons), improving usability and clarity.

2

Claims supported by cited references

Each recommendation is backed by named sources, allowing users to verify the rationale and explore more details independently.

Findings about the System

Finding 1

Recommendation about the finding

Sources: Source label with link Source label with link

Finding 2

Recommendation about the finding

Sources: Source label with link Source label with link

Interaction example: Suggested Actions

1

Consequences and benefits are explicit

Includes options with a clear summary of what it changes and what the effect will be, allowing the user to compare outcomes at a glance.

2

Multiple actions presented side-by-side

The interface surfaces more than one possible action instead of a single automated path. This supports user agency and accommodates different risk tolerances.

3

Supports informed trade-off decisions

By presenting pros and cons transparently, the system helps users make context-aware decisions, especially when no option is perfect.

4

Labels indicate duration and reversibility

Visual tags communicate whether an option is temporary, reversible, or long-term. This helps users understand not just the effect, but also the scope and risk.

Suggested actions to improve performance

First preferred action

Temporary measure Immediate, reversible

Measurable impact

Second preferred action

Medium-term measure 5 minutes

Measurable impact

How to implement

  • Include reasoning explanations alongside every recommendation or decision
  • Make explanations accessible through plain language and visual aids
  • Use progressive disclosure to offer both quick summaries and detailed explanations
  • Show alternative options considered and why they were not chosen
  • Provide clear source citations and links for verification
  • Display confidence levels and uncertainty ranges where relevant

Common pitfalls

Opaque decision logic

Users can't tell why the agent made a choice

Over explanation

Flooding users with too much technical detail and overwhelming them with too much detail

Recovery

Assume the agent will make mistakes, make them clearly fixable

?

What it means

Agents will make mistakes, what matters is how fixable they are. Recovery means giving users clear, safe ways to undo actions, correct errors, and guide future behavior. It makes systems feel less brittle and more collaborative.

!

Why this matters

Without recovery, even small errors can erode trust and stall progress. Clear ways to fix mistakes turn agent failures into moments of learning for both the system and the person using it.

Interaction example: Undo & Redo Support

1

Actions are reversible by default

The interface includes options like Undo or Revert for each automated change, giving users an immediate path to reverse agent decisions.

2

Justification for actions builds trust

A short, plain-language explanation helps users understand the rationale behind changes, reducing confusion and making recovery decisions easier.

3

Multiple levels of recovery available

Users can revert or approve individual changes or apply recovery to all actions at once. This offers both fine-grained control and bulk handling.

Agent Suggested Change

Based on recent activity the agent marked these tasks as complete and reassigned owners


Task Action
Draft Proposal
Marked as complete
Assign Reviewer
Reassigned to Alex

Why this happened

The agent detected inactivity for 5+ days and no blocking comments, so it marked the task as complete.

Interaction example: Editable Outputs

1

Human-in-the-loop decision making

The interface shows multiple alternatives, but waits for the user to select one. This maintains control and avoids premature execution.

2

Language supports co-creation

AI's phrasing encourages collaboration reinforcing the user as the final authority, not a passive observer.

3

Selected output is not final

Once the user picks an option, the system surfaces editable fields instead of applying the change directly. This allows precise customization.

Alternative 1

Explanation

Measurable impact

Alternative 2

Explanation

Measurable impact

Alternative 3

Explanation

Measurable impact

Let's go with alternative 1

A

Sure, before I go ahead with implementing this, please confirm the following details

Actionable 1 Metric in categories Category 1 for user
with exceptions for None

Interaction example: Safe Defaults

1

Builds trust through predictable, gradual control

Safe, consistent defaults help users gain confidence and expand control at their own pace.

2

Activation requires explicit user intent

Features that could affect security or behavior are opt-in only. This ensures users can explore safely and expand functionality on their terms.

Rules

Rules will ensure safe, conservative operation based on prior user interactions and best practices.

Default 1

Description of the default

Default 2

Description of the default

Interaction example: Escalation Paths

1

Manual input and escalation always available

Users can directly provide input or ask their own questions at any time, ensuring they remain in control and can override or guide the agent when needed.

2

Clear option to proceed independently

The "Alternate" option offers an immediate escape route — users can skip the assistant and take action themselves without friction or waiting for approval.

Important screen loaded by Agent

Agent asks a question about the screen because they cannot make a decision and need clarification from user

Describe step here

How to implement

  • Design agents to offer fallback options or manual alternatives instead of total failure
  • Use feedback from failure and recovery experiences to continuously improve system behavior
  • Make recovery options easy to find, context-sensitive, and layered from simple to advanced controls

Common pitfalls

Failing to learn from recovery events

Failing to analyze recovery patterns can lead to repeated mistakes, missing the opportunity to learn from user corrections and improve AI performance over time

Lack of granular control

Using only high-level revision features frustrates users who want to undo specific AI actions without losing their own work, missing the need for precise, collaborative recovery

Inconsistent recovery experiences

Recovery mechanisms that work differently across different parts of the system confuse users and create cognitive overhead

Unclear recovery guidance

Users need clear explanations and recovery options when things go wrong. Vague errors and unclear paths lead to frustration and reduced trust

Collaboration

Design for shared effort and mutual input

?

What it means

Autonomous agents should act as capable partners, not just tools waiting for commands. Collaboration means shared context, back-and-forth interaction, and joint ownership of outcomes. The agent contributes ideas, takes input, and improves the work in progress.

!

Why this matters

Collaboration builds stronger results than automation alone. When people and agents shape outcomes together, users stay engaged and push toward more creative, effective solutions.

Interaction example: Mixed Initiative

1

Agent proactively initiates based on context

Allow the agent to proactively detect issues or make suggestions, especially when it has useful context the user may not.

2

Build on each other's contributions

Design interactions so the agent can refine its outputs in response to human edits or questions. Keep the flow continuous and collaborative.

Incidents Detected by Agent

🔴 Incident 1

Description

🔴 Incident 2

Description

🟠 Incident 3

Description

🟠 Incident 4

Description

Triage Incident

A

I have detected a major issue starting at 09:17 UTC. What would you like to do?

Compare it to yesterday, and exclude issue x from the analysis.

A

Done. Compare to yesterday's baseline, it shows a 30% decrease in traffic. I recommend rollback. Would you like to rollback.

No rollback yet. Pull logs from first — I want to check status.

Interaction example: Co-editing Interfaces

1

Keep AI suggestions non-intrusive

Present changes as proposals, not automatic edits. Let users review, accept, modify, or reject.

2

Work in shared view

Both AI and human should operate on the same content in the same workspace. Transparency builds clarity and trust.

3

User always has final say

The human is the editor-in-chief. AI assists, but never publishes or commits changes on its own.

4

Make editing modalities clear

Let users choose how the AI helps — proofreading, rewriting, suggesting changes, etc. Provide flexible control modes, not just one-size-fits-all.

Policy Editing

Version: 1.2 Owner: team Name Last Edited By: Jane Rivera

Policy Details   Editing

Objective

A long description of policy details document added here by the human user.

Scope

Applies to all employees, contractors, and third-party users accessing internal systems via the corporate VPN.

Policy Rules

1. Rule 1 for Policy

Description of the rule written by the human. This is the rule selected and visually distinct from rest.

2. Rule 2 for policy

Description of rule

3. Rule 3 for policy

Description of rule

4. Rule 4 for policy

Description of rule

Assist

A

Suggestion:

"This is the recommended text for policy rule number 1"

Ask the AI Assistant a question

Assistant can make mistakes. Verify responses.

Interaction example: Role Clarity & Turn Signals

1

Clear stage-based ownership

Break workflows into visible stages and indicate who leads each one. This reduces ambiguity and improves accountability.

2

Attribute every action

Label actions clearly as system-initiated or human-initiated. This helps users interpret intent and trust the flow.

3

Signal when it's the user's turn

Use prompts, buttons, or callouts to indicate when the system is waiting for user input. Avoid passive steps that could confuse ownership.

Agent action review required

Waiting for review

Status

AI Drafts Changes
2
Engineer Review
3
Peer Approval
4
Staging Deployment
5
Commit

Description

Description of the agent action

Agent Rationale

  • Reason 1
  • Reason 2
  • Reason 3

Actions

This change is pending your review. It will not move forward until approved. Upon confirmation, it will proceed to peer review for secondary validation.

How to implement

  • Define clear, intuitive ways for control to shift between human and AI
  • Let the AI learn from collaborative successes and adapt its behavior to match individual user styles and preferences
  • Maintain a unified workspace that tracks contributions, context, and progress from both human and AI participants
  • Gracefully handle simultaneous edits with merging, version comparison, or deferring to human review when needed

Common pitfalls

Lack of transparency

Users don't understand how to influence the agent or override agent actions

Assumed alignment

The agent acts without confirming intent or context

Rigid flows

The system doesn't adapt when users try to collaborate or redirect

Binary choices

Only offering accept/reject rather than co-create options

Traceability

Make agent behavior visible, searchable & open to review

?

What it means

Traceability ensures agent decisions can be reviewed, understood, and improved over time. It makes behavior accountable across sessions, users, and workflows supporting debugging, learning, and workflow improvements.

!

Why this matters

As agents evolve, so do their decisions. Traceability allows teams to track changes, understand outcomes, and stay aligned in multi-user environments. It turns opaque processes into something you can audit, learn from, and improve.

Interaction example: Action History

1

Make events time-stamped and ordered

List all system and human actions in a clear sequence. Timestamps build trust and help reconstruct events during audits or investigations.

2

Include cause and effect where possible

Show how one step led to the next. This helps users understand the rationale and logic behind changes.

3

Capture both automated and manual steps

Record not just user input, but also system decisions. A full picture improves transparency and enables accountability across the workflow.

4

Use clear, plain language

Write log entries in simple, readable terms — no code dumps or vague system jargon. Everyone should be able to follow what happened.

Failure Log

1:45 PM: Issue detected at this time by the Agent

Agent Action

2:00 PM: Agent flagged to the user for optimization

Agent Action

2:15 PM: Resolution by the agent, and displayed on UI

Agent Action

2:30 PM: User reviews the issue

Reviewed by John Doe

2:40 PM: User approves the proposed resolution

Approved by John Doe

Interaction example: Visual Diffing

1

Use side-by-side comparisons

Display the original and updated states in parallel columns. This helps users spot differences immediately without extra mental effort.

2

Include the why, not just the what

Pair the visual change with a short explanation of the reason or logic behind it. This gives context and supports better decision-making.

3

Highlight what changed

Use color or styling to draw attention to fields or values that were modified. Don't make users guess what's different.

4

Let the user validate or intervene

Offer a clear way to accept, reject, or adjust the change. Visual diffs should inform action, not just display information.

Review Changes

Due to issues agent changed default value from old value to new value

Before

After

Interaction example: Behavior Tuning Over Time

1

Call out what triggered the change

Clearly state the condition or threshold that caused the system to respond differently than before.

2

Compare past vs. present behavior

Provide users a way to see what's new vs. what used to happen. This helps them understand system learning and decide if further intervention is needed.

3

Explain the system's current decision logic

Let users understand why the system acted in this instance and how it may influence future behavior. Indicate whether this reflects a one-time response or an evolving pattern.

4

Allow control or rollback

Include an option to undo, override, or adjust the system's adaptive behavior.

Routing Changes

Why this happened

Agent based on observation modified behavior of the system

Details

Where change happened Description What change happened Description New Solution Details of Solution (active since 2:30 PM) Previous Solution Details of Solution
Reasoning for choosing this path
How it worked previously

How to implement

  • Make it easy to trace outputs back to the inputs, prompts, or interactions that influenced them
  • Provide interfaces that let users review, filter, and explore past actions and decisions in a structured, searchable way
  • Record all system and AI actions with timestamps, inputs, outputs, and relevant context to support clear trace trails
  • Ensure the system's behavior can be independently reviewed and traced to support transparency and hold the system accountable

Common pitfalls

False consistency

The system behaves differently in similar situations

No feedback loop

Users don't see whether the action succeeded or failed

Build with the Hax SDK

The HAX SDK gives developers everything they need to integrate agents into their apps, without losing clarity, structure, or control. Use structured schemas, prebuilt components, and clear boundaries to keep agent behavior collaborative and predictable.

Explore the SDK
Hax SDKHax SDK