shell + probeBusiness and Personal processes running; no recent app crashes; plenty of free disk space.
RAPPORT AGENT WORKSPACE
Why won’t
my files sync?
“The app is running. Let’s check what’s happening underneath.”
Talk to your device. Cloud AI reasons through the problem; a local harness gathers evidence and carries out permitted work. You guide the investigation, review held actions and check what actually changed.
Explore edited walkthroughs from internal use—sync, sleep, unsaved work and recent changes. No account needed.
Agent Workspace01 / 04Look beneath the symptom
OneDrive is running. Both accounts report no upload errors and no queued work. That alone doesn’t explain why a file isn’t arriving.
I’m checking the account logs, sync folders and macOS File Provider layer together.
Inspect the supporting evidence 2 checks
read_fileBoth accounts report zero upload/download errors and an empty queue.
An empty error queue is evidence to investigate, not proof that sync is working.
Files won’t sync · step 1 of 4Your people. Their devices.
Finally on speaking terms.
Ask the deviceStart with the problem
Review the actionKeep control of the change
Check the resultContinue the conversation
CLOUD INTELLIGENCE. DEVICE CAPABILITY. HUMAN DIRECTION.“Intelligence in the cloud.
“Intelligence in the cloud.
Hands on the device.
A person in between.”
Rapport pairs cloud AI with a working harness on each enrolled device. The harness is the local software that manages the tool loop, checks proposed commands and runs permitted tools. You work with both through one conversation.
Reason in the cloud.
The model interprets the evidence and proposes the next check or action. Each turn goes through Rapport’s gateway, with tenant model settings, session limits and usage records.
Model credentials stay server-side.Direct the work.
Describe the problem. Challenge a hypothesis. Inspect the evidence and approve or decline held commands. Keep the investigation connected to the person experiencing the issue.
Your question. Your judgment. Your next step.Work where the issue is.
The native agent reads device state, gathers diagnostic evidence and executes permitted tools. It classifies shell commands locally before execution; the model’s request alone is not permission.
Local execution. Policy checked on the endpoint.ONE WORKING LOOP
Your question Cloud reasoning Local policy check Permitted device work Evidence for the next turn
Allowed read-only checks can run automatically. Mutating or elevated shell commands are held for approval. Commands classified as destructive or disallowed are refused.From seeing an issue to working through it.
Experience scores and fleet signals tell you where to look. The workspace lets you investigate the next question on that device, gather fresh evidence and review a tailored action. Enabled tools can address a question without a prewritten workflow for that exact issue.
Cloud reasoning uses device evidence.
Conversation context and bounded tool results travel through the gateway to the configured AI provider. Output scrubbing reduces exposure; it does not guarantee that all sensitive data stays on the device.
Explore execution and data boundariesTHE AGENT WORKSPACE IS THE STARTING POINT“Show me
“Show me
what’s going on.”
The agent runs tools on the selected endpoint as it investigates. The workspace keeps your question, its checks, proposed actions and returned results together.
- 01
Ask in your own words.
“Why is this device slow?” Start with the problem. Continue with follow-up questions in the same conversation.
- 02
See why it answered that way.
Inspect device checks and tool output alongside the explanation. Keep an observation separate from a suspected cause.
- 03
Review the proposed fix.
Inspect the exact command and target when an action needs approval. Approve or decline that request in the workspace.
- 04
Ask it to check again.
Review the execution result and fresh device evidence. A completed command is not the same as a confirmed resolution.
macOS · Sync investigationRecorded readbacks
- Business sync
- Paused → active
- Personal sync
- Remained active
RAPPORT’S EXPLANATION
“The Business account is active again. Personal is still healthy. Please check that the original file reached its destination.”
CONVERSATIONAL SUPPORT. EXPLICIT BOUNDARIES.“Let me review
“Let me review
that first.”
The conversation can propose a device change. The workspace shows the request for a person to review; a chat message is not a substitute for the required approval.
See the workspace’s security boundariesThe right operator, in the right tenant.
Tenant-scoped access and MFA-backed operator sessions.
Checks within configured limits.
Enabled tools, session limits and bounded results—not unrestricted access.
A command you can inspect.
Review the requested action and its target. Approve or decline the individual request.
A record of the conversation.
Follow the messages, tool calls, approval decisions and returned results.
THE PLATFORM BEHIND THE CONVERSATION“What’s behind
“What’s behind
the answer?”
The workspace leads. Device context, diagnostic tools, controlled actions and fleet operations support the core experience: working through an issue with the device.
Digital experience
Experience scores and telemetry help you choose where to start the conversation.
Performance · stability · connectivityAI-assisted investigations
Diagnostic observations help the assistant explain what it found and what to check next.
Observations · findings · next checksGuided remediation
Supported actions turn a conversation into a reviewable next step on the device.
Target · proposed action · returned resultPatch orchestration
When the next step is a wider change, stage the rollout and review each ring.
Pilot rings · health evidence · rollout gatesAgent release controls
Keep the endpoint agent itself under your tenant’s release and adoption policy.
Approval · automatic adoption · holdGovernance and trust
Collection policy and access controls define the boundaries around the work.
Collection scope · operator access · recordsWant to inspect a fleet-level decision?Explore the supporting workflow demos
START WITH ONE DEVICE AND ONE REAL QUESTION“Let’s start
“Let’s start
with one device.”
Enable the Agent Workspace for an agreed pilot cohort. Confirm the tools, operator access and AI data scope, then work through a representative issue together.
Plan your evaluationBEFORE THE FIRST CONVERSATION“A few questions
“A few questions
before we start.”
What is Rapport?+
Rapport pairs cloud AI with a local device harness in a conversational workspace for IT. The harness manages the tool loop and permitted execution on each enrolled device. Ask a device why it is slow: the agent runs diagnostic tools on demand on that endpoint, explains what it finds and proposes the next step. Inspect the results, review a proposed action and continue the conversation. Device intelligence, diagnostics, remediation and fleet controls support that central Agent Workspace experience.
Is this a live AI conversation on the public website?+
No. The Agent Workspace walkthroughs condense real internal sessions, with identities removed and wording edited for clarity. Their controls navigate recorded examples; they do not approve or execute anything. Supporting fleet demos use fictional data. In a configured deployment, the endpoint agent runs the tools and the gateway calls your configured AI provider.
What keeps a conversation from becoming an uncontrolled change?+
Workspace access requires a tenant operator with an MFA-verified session. Enabled tools and configured limits govern the work; held commands are shown for an approve or decline decision. Approval applies to the individual requested command, not every later action. The agent is a privileged service, so these controls do not replace careful scoping of operator access.
Does the AI run on the device?+
No. The model runs in the cloud, reached through Rapport’s governed gateway. The native device harness manages the tool loop, checks shell commands against local policy and executes permitted tools on the endpoint. Prompts, conversation context and bounded tool results go through the Rapport gateway to the configured AI provider, and conversation history is stored in the backend. Provider API keys remain server-side. Review data scope, retention and provider terms before enabling the workspace.
Can we start with a pilot?+
Yes. Rapport is at the beta and pilot stage. Contact the team to discuss a focused evaluation with agreed devices, collection scope, support workflows and success criteria before considering wider deployment.
Which devices can we evaluate?+
The agent supports eligible Windows, macOS and Linux systems, with Windows MSI and macOS PKG deployment paths. Measurements and actions vary by operating system and configuration. The pilot should confirm your specific OS versions, hardware and device-management requirements.
How does Rapport approach employee privacy?+
Experience collection is configured by policy and excludes employee keystrokes, screen capture, message content and browsing history. It does not score individual productivity. Aggregate workforce analytics have separate access controls and minimum-group thresholds. Authorized support actions can return additional device information or selected files, so those permissions and collection notices should be agreed separately.
Does an AI finding automatically change a device?+
No. AI-assisted investigations use read-only diagnostics to assemble observations and recommend next steps. Device changes follow the enabled endpoint policy and applicable approval and verification requirements. A finding's confidence is an aid to review, not a guarantee of root cause or a permission to execute.
Can we decide when changes reach our fleet?+
Yes. Patch deployments use maintenance windows, staged rings and evidence gates. Agent releases have a separate choice of individual approval, explicit automatic adoption or hold, with rollout and timing controls. A hold stops future eligible update attempts; an installation already in progress may finish. Device-management administrators still control packages they deploy directly.
“Why is this device slow?”
Let’s work through it.
Create your organization, secure your account, and invite your team. Plan your first device rollout with us.
Rapport is available for beta and pilot evaluation. Workspace enablement and AI configuration are agreed during the pilot.