Use agent progress trace for live multi-step agent execution after work starts; agent plan preview before execution; streaming response for generated output chunks; activity log for retained evidence; progress bar for measurable bounded work; and human approval gate for one paused authorization checkpoint.
Use agent progress trace for live multi-step agent execution after work starts; agent plan preview before execution; streaming response for generated output chunks; activity log for retained evidence; progress bar for measurable bounded work; and human approval gate for one paused authorization checkpoint.
Practice the judgment: Imagine this situation: A multi-step agent run shows only a vague spinner or Thinking label. Write the person's goal, choose one of these patterns, and show the state that helps the person continue. Use the decision rules and source trail below to explain your choice.
Review the underlying principle →| Dimension | Agent progress trace | Agent plan preview | Streaming response | Activity log | Progress bar | Human approval gate |
|---|---|---|---|---|---|---|
| UI or UX | UI + UX - Live execution trace for an AI agent or automation run after work has started | UI + UX - Pre-execution preview of an AI agent's proposed multi-step plan, tools, data access, and expected outputs | UI + UX - Incremental generated-output surface that renders response chunks before final completion | UI + UX - Searchable and exportable record of system, user, or administrative events | UI + UX - Measurable system-operation progress indicator | UI + UX - Runtime checkpoint that pauses AI or automation until an eligible human authorizes the next step |
| UI guidance | Render an agent progress trace as a live, ordered run timeline with run ID, plan version, current step, queued steps, active tool or task, elapsed time, last event time, step status, blocked gates, retry state, and final outcome. | Render agent plan preview as a pre-run plan with objective, ordered steps, planned tools, data sources, permissions, assumptions, dependencies, approval gates, expected outputs, and controls to edit, approve, run, save, or cancel. | Render streamed output with a visible generation state, partial-answer label, stop control, final-complete state, and clear distinction between text that is still arriving and content that has passed final citation, safety, tool, or format checks. | Render activity logs as evidence-oriented records with event time, actor, action, object, source system, scope, result, and technical context such as IP address or location when available. | Show a labeled bar with a track, filled value, and nearby helper text that reports the measurable unit such as percent, bytes, rows, files, records, or stages. | Render a human approval gate as a paused automation checkpoint with the proposed action, tool or workflow step, triggering rule, risk level, payload snapshot, requester or agent, approver eligibility, timeout, and explicit approve, reject, edit, cancel, or bypass controls. |
| UX guidance | Use agent progress trace when an AI agent or automation has started multi-step work and users need to monitor progress, intervene on stalls or gates, understand partial completion, and know whether the reviewed plan is still being followed. | Use agent plan preview when users need to understand and shape what an AI agent will do before it starts calling tools, changing records, sending messages, spending budget, or making external side effects. | Use a streaming response when showing partial generated output helps users start reading or monitoring work before the model finishes, and when the product can explain that early chunks may still change, be filtered, or lack final sources. | Use activity log when users need to investigate, audit, verify, or troubleshoot actions across accounts, objects, systems, settings, or security boundaries. | Use a progress bar when the system can honestly report movement toward a known finish and users need to decide whether to wait, cancel, retry, continue elsewhere, or return later. | Use human approval gate when automation is ready to act but policy, risk, confidence, cost, access, publication, deployment, customer impact, or legal consequence requires a human decision before execution continues. |
| Good UI | An account research agent trace shows Run A-204, reviewed plan P-18, completed CRM lookup, active policy search, queued draft email, approval gate pending before send, elapsed time, and a View tool details control. | A sales assistant previews a six-step account-research plan with CRM lookup, web search, draft email, approval gate before send, estimated sources, and editable recipient scope. | A policy assistant shows Answer generating, streams paragraphs into a stable answer region, marks citations pending, exposes Stop generation, then changes to Complete when citations and safety checks finish. | An organization audit log table shows timestamp, actor, action, target object, app, IP address, result, and a Details drawer with before and after fields. | A document upload card says Uploading evidence.zip, shows 64%, 32 of 50 MB, a Cancel action, and keeps the rest of the form usable. | An AI support agent pauses before issuing a refund, shows the proposed amount, customer, policy match, confidence, source grounding, approver role, timeout, Approve refund, Edit amount, Reject, and Stop run controls. |
| Bad UI | A trace displays raw tool JSON and private tokens as the only progress feedback. | A list of completed tool events is titled Plan preview, confusing proposed work with already executed work. | A streaming JSON answer is displayed as valid data before the closing braces and schema check have completed. | An export button downloads a different date range and timezone than the on-screen filtered log. | A progress bar reads 100% complete while virus scanning and server acceptance are still pending. | A reviewer can approve an automation run after the action already executed, making the gate only decorative audit text. |
| Good UX | A user watches the active step move from searching policies to drafting the email, opens the blocked permission item, grants access, and sees the run continue from the same step. | A manager removes the Send email step, narrows the data source to approved knowledge, approves the remaining plan, and sees execution start from the revised version. | A user sees the first-token state quickly, reads early outline bullets while the answer continues, stops generation after enough detail, and sees the result labelled Partial with Continue and Regenerate options. | An admin filters to failed SSO events, expands one entry, copies the event ID, exports the filtered range, and sees that records older than 180 days require a different archive. | A user uploads evidence.zip, sees progress move from 12% to 64%, cancels before commit, retries after a network error, and gets a completed receipt only after server processing succeeds. | A billing lead opens the paused refund gate, sees that the amount is under policy but source grounding is partial, edits the refund to the verified amount, approves, and the agent resumes only that step. |
| Bad UX | A trace restarts from the top after a retry and makes users think earlier completed steps ran twice. | Users approve a plan that says Research account but the agent also updates the opportunity stage. | A user copies an early legal recommendation before the final paragraph reverses the conclusion after a tool result arrives. | A user marks a notification read and the corresponding activity evidence disappears from the only log. | Users leave after 100% transfer even though server processing later rejects the file. | A human approves a stale agent action from email and the agent applies it to a different customer state. |
| Best fit | An agent or automation run has started and spans multiple steps, tools, gates, or side effects. | An AI agent or automation can show a proposed multi-step plan before execution. | Generated text or structured content can be read or monitored before completion. | Users need to inspect recorded user, admin, system, security, or integration events. | A system operation has a measurable total or bounded progress value. | An AI agent, workflow, deployment, or automation is ready to perform a high-impact step and must pause for human authorization. |
| Avoid when | The task is a single generated answer stream. | The work is a simple single recommended action. | Intermediate chunks may expose unsafe, private, or misleading content. | The goal is only to show a readable milestone history for one case or process. | The wait is short and only needs scoped activity feedback. | The action has already happened and users only need an audit log. |
| Required state | Run started state tied to run ID, plan version, objective, and user who started the run. | Draft plan state with objective, ordered steps, planned tools, and expected output. | Queued or receiving state before first output arrives. | Default log state with event records, result count, visible timezone, retention window, and permission scope. | Idle state before the operation starts. | Paused gate state with proposed action, payload snapshot, reason for gate, and run context. |
| Accessibility burden | Expose trace status, run ID, current step, elapsed time, blocked state, final outcome, and details availability as text. | Expose objective, plan version, step order, step status, tool, data access, side effect, and expected output as text. | Expose stream milestones such as started, still generating, stopped, failed, citation ready, and complete as status messages. | Use table or structured list semantics so actor, action, object, timestamp, result, and scope are perceivable together. | Provide an accessible name that identifies the operation and affected object. | Expose gate status, proposed action, target, payload summary, risk, approver rule, timeout, and current run state as text. |
| Common misuse | Using one spinner or vague Thinking label for a multi-step agent run. | Showing a vague plan summary while hiding planned tool calls, data access, and side effects. | Letting partial output look finished before the final event arrives. | Using vague activity labels that omit actor, action, object, timestamp, or result. | Fabricating progress values just to make users feel movement. | Using a post-action audit record as if it were a pre-action human gate. |