The complete MS Duplex capability list.
114 capabilities in 17 groups, from the context every task starts with to an advantage that compounds with every task.
Together, they make MS Duplex the Agent Operating System powered by Runtime Intelligence: the layer every agent runs on, and one that learns. Why it is one →
17 groups, 114 capabilities.
A dot for each capability. Press a group to open it.
- Context and observation
- Rules and behavioural control
- Tool execution
- Learning and personalisation
- Long Horizon: long-running work
- Planning and decision support
- Skill Fabric: reusable skills
- Persistent threads and sharing
- Forgetting and data control
- Status and explanations
- Persistent intelligence
- Personalisation
- Multimodal intelligence
- Confidence and autonomy
- Runtime safety and oversight
- Operational intelligence
- Integration and portability
Context and observation
7Every task starts with what matters and with the same, current facts, and every finished task leaves context for the next.
1Relevant context for every task
MS Duplex brings in the past experience, documents, rules, skills, conversations and media that matter to the work at hand.
Your agents start every task informed, instead of from a blank page.
See it at work →2Learning from finished work
MS Duplex learns from what an agent answered, which tools it used and what happened as a result.
The next agent on similar work starts with what this one found out.
See it at work →4Facts checked when they are used
Where a decision depends on a fact, MS Duplex reads its current value from its source when the decision is made.
Decisions use today’s value, not a remembered one.
See it at work →5Facts changed by the right people
Only the people entitled to change a fact can change it, and every change is on record.
Your facts stay trustworthy as they change.
6New facts proposed from the work
When work turns up a new or changed fact, MS Duplex proposes it to the people who own it.
Your facts stay current without anyone hunting for changes.
7A history for every fact
Every fact keeps its history: what it was, when it changed and who changed it.
Any decision can be traced back to the facts it used.
Rules and behavioural control
3Your rules are visible, written with you and changed only once you confirm.
8Active rule discovery
MS Duplex shows every personal, team, workspace and organisation-wide rule governing the current work.
An agent knows the limits of the work before it acts, and so does your team.
See it at work →9Rule creation
MS Duplex turns a policy you describe into a rule your agents follow, and shows you exactly what it means before you confirm.
A policy your team agrees on becomes how every agent behaves.
See it at work →10Rule removal
MS Duplex removes an outdated rule once you confirm exactly what will change.
Your rules stay current, and no rule changes by accident or by misunderstanding.
Tool execution
1Your business tools run once the action is allowed.
11Approved tool execution
MS Duplex runs registered business tools only after checking the action against permissions, rules and the current task.
Your AI goes from giving advice to doing the work, such as issuing a refund, within the limits you set.
See it at work →
Learning and personalisation
10MS Duplex learns from examples, facts, documents, corrections and outcomes while the work happens.
12Label learning
MS Duplex learns how your work should be sorted from examples, corrections and the terms your organisation already uses.
Sorting gets faster and more consistent, and matches the way your organisation works.
See it at work →13Fact learning
MS Duplex learns the facts that matter to your work and brings them up whenever they become relevant.
A useful fact outlives the conversation it came up in.
See it at work →14Document learning
MS Duplex learns from your manuals, policies, reports, runbooks and reference files, and draws on them as the work happens.
The right page of the right manual reaches an agent at the moment it helps.
See it at work →15Learning what actions do
MS Duplex learns what each decision or action leads to in your systems.
Agents understand how your work moves from one step to the next, and what each step changes.
See it at work →16Tool-use learning
MS Duplex learns which tools work best for different situations and how those tools should be used.
What worked with a tool once helps every agent that needs it next.
See it at work →17Credit to the right decisions
MS Duplex identifies which earlier decisions and actions contributed most to a final result.
The steps that made the difference count most, instead of every step counting the same.
See it at work →18Outcome learning
MS Duplex learns whether work succeeded, failed or landed somewhere in between, and lets that shape the next decision.
What happens in the real world keeps improving how the work is done.
See it at work →19Correction learning
MS Duplex turns a corrected answer, label or action into better behaviour for the future.
MS Duplex makes feedback immediately useful without waiting for retraining or a new model release.
See it at work →20Method validation
MS Duplex checks whether a reusable procedure works each time it runs, and learns from every attempt.
A method that keeps working earns more trust each time it is used.
See it at work →21Skill validation
MS Duplex verifies whether a learned skill produced the intended result in real work.
A skill earns its place by working, and loses it when it stops.
See it at work →
Long Horizon: long-running work
15Work that lasts days or months keeps its goal, its plan, its progress and its safe points, through pauses, hand-offs and change, until the goal is reached.
22Persistent task creation
MS Duplex makes long-running work one task, with its own goal and progress, that outlives any single conversation.
Complex work can run for days, weeks or months and still know where it is heading.
See it at work →23Task progress tracking
MS Duplex tracks each important action, change, result and check as the work moves forward.
Anyone can see what has happened so far, and what should happen next.
See it at work →24Task checkpointing
MS Duplex marks a safe point in the task before a big or risky change is tried.
You can trust agents with bolder work, knowing there is a safe point to go back to.
See it at work →25Task rollback
MS Duplex takes the task back to an earlier safe point when a later approach fails or has to be dropped.
Agents can try a new approach without risking the work done so far.
See it at work →26Task-state inspection
MS Duplex shows where a task stands, how far it has come and what is next, whenever you ask.
Anyone can pick up a long job without digging through old conversations to piece it together.
See it at work →27Task completion
MS Duplex closes a task with a final check, and learns from its outcome and the lessons along the way.
Every finished job makes the next similar one easier for your whole organisation.
See it at work →28Task discovery
MS Duplex lists every active, paused, finished and retained task in one place.
Important work stays in sight, instead of buried in old chat histories.
See it at work →29Task handover
MS Duplex brings together the goal, the decisions so far, the progress, the open questions and the next steps, so a task can carry on.
Another session, agent, app or model picks up the work where it stopped.
See it at work →30Plans checked step by step
Before each step of a long plan runs, MS Duplex checks that what the step depends on still holds.
Each step of a long job runs on today’s facts, not last week’s.
31Plans revised when things change
When a step no longer fits the situation, MS Duplex revises the rest of the plan from the work already verified.
Long work adapts to change instead of starting over.
32Approved exceptions
When a step has to go beyond the plan or your rules, the people you choose approve the exception, and it is on record.
Your team stays in charge of long work, even when the plan changes.
33Verified progress
A step counts towards the goal only once its result has been checked.
Long work cannot look finished before it is.
34Stalled work caught early
MS Duplex notices when long work stops moving towards its goal, goes round in circles or runs past the budget you set.
A stuck job is spotted early, while there is still time to change course.
35Work taken up again
After a pause or a hand-off, MS Duplex takes the work up where it stopped, with its goal, plan, decisions and approvals.
Long work carries on through the interruptions real work is full of.
36Lessons from the path that worked
When a task closes, what was learned along approaches that were dropped is withdrawn, and the steps that led to the result are credited.
Only what led to the result shapes the next job.
Planning and decision support
6MS Duplex looks ahead before anything is committed, and recommends the next move.
37Intelligent classification
MS Duplex sorts a new case into the categories it has learned, and knows how sure it can be of the answer.
Similar cases are sorted the same way every time, and an unclear case is never forced into the wrong box.
See it at work →38Agreement across multiple inputs
MS Duplex compares different wordings, observations or candidate answers and settles on the best-supported one.
When inputs repeat or contradict each other, you get one consistent answer instead of several.
See it at work →39What an action will do
MS Duplex predicts what a proposed action will change before anyone takes it.
Agents catch avoidable mistakes before they happen, and choose the steps that get the work done.
See it at work →40Scenario simulation
MS Duplex plays a possible sequence of actions forward, step by step, to see how it would unfold.
You see the likely consequences, risks and alternatives before any time or money is spent.
See it at work →41Plan construction
MS Duplex turns a goal into a step-by-step plan, drawing on the context, your rules and what it has learned about your systems.
Plans follow how your work behaves in practice, not a generic checklist.
See it at work →42Action recommendation
MS Duplex ranks the available next actions and identifies the option best suited to the current situation.
Agents take each next step for a reason, instead of by trial and error.
See it at work →
Skill Fabric: reusable skills
10What worked once becomes a skill your agents can find, inspect and trust, shared as far as you choose and kept current as your work changes.
43Skill creation
MS Duplex turns work that has succeeded, and been checked, into a skill your agents can use again.
One good solution goes on to serve many future tasks.
See it at work →44Skill recall
MS Duplex brings the best-suited skill it has learned to the task at hand.
An agent gets the right proven skill exactly when it needs one.
See it at work →45Skill discovery
MS Duplex lists the skills available in the current workspace.
Your team sees what its agents can already do, instead of working it out again.
See it at work →46Skill inspection
MS Duplex shows what a skill does, when to use it and the knowledge it draws on.
Anyone can open a skill and understand it, so nothing your agents use is a black box.
See it at work →47Skill track record
MS Duplex shows how a learned skill was tested and how it has performed since.
Your team can tell a skill that has proved itself from one nobody has tested yet.
See it at work →48What a skill is made of
MS Duplex shows the knowledge and learned steps that make a skill work.
Complex skills become easier to judge, improve and trust.
See it at work →49Skills shared across your organisation
A skill one team has proved can be published to other teams and to the whole organisation, once it is approved.
Good work spreads through the organisation instead of being rebuilt team by team.
50Skills watched in use
MS Duplex follows how each skill performs in real work, uses the ones that keep delivering and retires one that stops.
Your skills stay current as your work and your systems change.
See it at work →51Skills from long work
A skill learned during a long job is kept once the job’s result is verified.
Your skills grow from work that reached its goal.
52Skills that outlast your model
Skills live in MS Duplex, not in the model, so they stay when you change models, clients or vendors.
Your skills are yours, whichever model you rent.
Persistent threads and sharing
5Ongoing work has one home across sessions, shared only on your terms.
53Thread discovery
MS Duplex lists the persistent work threads carried forward from previous sessions and projects.
MS Duplex makes ongoing context easy to find whenever related work returns.
54Thread inspection
MS Duplex brings up the key events, decisions and context from a thread.
You can always tell trusted context from what someone else shared.
See it at work →55Persistent thread creation
MS Duplex opens a thread for a project, a responsibility, a client, a workflow or any work that carries on over time.
Related work has one home you can rely on, across sessions and agents.
56Controlled thread sharing
MS Duplex gives an approved person or team access to one thread, and nothing beyond it.
People can work together on a thread while everything private stays private.
Forgetting and data control
8Anything MS Duplex has learned can be removed, from one label to a whole task.
58Label forgetting
MS Duplex forgets an outdated or wrong label, and stops using it to sort new work.
Your categories can change when your business does, with nothing to retrain.
See it at work →59Correction forgetting
MS Duplex undoes a correction that no longer applies, and the earlier behaviour returns.
What MS Duplex has learned can change as your requirements and preferences do.
See it at work →60Thread forgetting
MS Duplex removes a persistent work thread and the context carried through it.
Sensitive or outdated work can be retired for good, instead of lingering forever.
61Task forgetting
MS Duplex removes a task, and everything needed to resume it, when the work should no longer exist.
Long-running work gets a clean ending, on your terms.
See it at work →62Credit forgetting
MS Duplex removes a wrong link between a result and the actions given credit for it.
A lesson drawn from the wrong cause stops steering future decisions.
See it at work →63Outcome forgetting
MS Duplex withdraws a chosen success or failure, along with what it taught.
A misleading, outdated or inappropriate result no longer shapes how the work is done.
See it at work →64Observation forgetting
MS Duplex forgets a wrong observation about what an action leads to, and checks the plans built on it.
Plans follow how your workflows run today, not how they once ran.
See it at work →65Example forgetting
MS Duplex removes selected examples that taught it a behaviour or a skill.
You choose exactly which examples shape how your agents behave.
See it at work →
Status and explanations
11You can see what MS Duplex did, why it did it and what supports it.
66Decision auditing
MS Duplex shows which context, rules, examples and actions led to a result, and how sure it was.
Anything an agent did can be reviewed, explained and corrected.
See it at work →67Activity history
MS Duplex shows a timeline of the important actions, outcomes, lessons learned, approvals and rule changes.
Your team can see what happened across every agent and session, long after the fact.
See it at work →68Job tracking
MS Duplex reports whether longer-running work is waiting, active, completed or unsuccessful.
You can see background work at a glance, without checking on it again and again.
See it at work →69Method inspection
MS Duplex shows the current state, health, history and observed results of a reusable procedure.
MS Duplex makes it clear which operating methods are dependable and which ones need improvement.
See it at work →70Capability discovery
MS Duplex shows every action, connected tool and permission available in the current workspace.
A new agent knows straight away what it can ask for, and uses MS Duplex correctly from the start.
See it at work →71Policy visibility
MS Duplex shows the settings in force for context, confidence, speed, adaptation and decisions.
Your team can see how each work area is set up, and why it behaves the way it does.
See it at work →72Checks behind every change
MS Duplex shows how each verified piece of knowledge, skill, decision and reversal was checked.
Trust rests on checks your team can see, not on someone’s word.
See it at work →73Skills built from experience
MS Duplex shows the skills it has built from work that succeeded and was checked.
Your team can open any of them, check it and use it again.
See it at work →74Method discovery
MS Duplex lists the reusable procedures learned from repeated work and real-world outcomes.
What keeps working becomes a library of proven methods your team can see.
See it at work →75Skills shaped to your work
MS Duplex shows the skills it has shaped to the knowledge, tasks and way of working in the current workspace.
You can watch your AI fit your work more closely, one skill at a time.
See it at work →76Knowledge-source discovery
MS Duplex shows which approved sources a work area can draw on for its answers and decisions.
Anyone can check where an agent’s knowledge comes from before trusting its answers.
See it at work →
Persistent intelligence
6What MS Duplex has learned outlasts the session, the model and the app.
77Cross-session continuity
MS Duplex carries what it learned, and where each task stands, past the end of a conversation.
Every new session picks up where the last one stopped, so nobody repeats themselves.
See it at work →78Cross-model continuity
MS Duplex carries its knowledge, skills, rules, tasks and outcomes over to any new model you choose.
You can change model provider and lose none of what your AI has learned.
See it at work →79Cross-application continuity
MS Duplex carries the context of your work as it moves between supported AI apps.
What one app or agent learned is ready for the next one you use.
See it at work →80Agent-to-agent handoff
MS Duplex prepares the focused context another agent needs to continue the same work.
Several agents work like one, instead of starting over at every handover.
See it at work →81Context sized to fit
MS Duplex condenses a long history to the facts, decisions and open work that matter for what comes next.
Your model reads what the next step needs, instead of the whole transcript.
See it at work →82Continual learning without retraining
MS Duplex applies new facts, corrections, outcomes, workflows and skills without changing the underlying model’s weights.
Your AI improves from the next request, without the wait and cost of fine-tuning.
See it at work →
Personalisation
7MS Duplex fits the person, the team and the organisation, and what one of them proves can serve the others.
83Individual personalisation
MS Duplex learns each person’s preferences, terms, projects, corrections and way of working.
Each person’s AI feels more like their own over time, and less like a generic assistant.
84Team personalisation
MS Duplex turns the knowledge, skills, methods and practices your team approves into intelligence the whole team shares.
Every agent starts with what the team has already learned.
See it at work →85Organisation-wide personalisation
MS Duplex carries company knowledge, standards, experience and proven practice across the whole organisation.
What your organisation knows outlasts changes of people, apps, agents and model providers.
86From one person to the team
What one person’s agents have proved can be promoted to the team or the organisation, once it is approved.
Your best practice becomes everyone’s, with your say.
87Every member starts informed
A new member’s agents start with what their team and organisation have published.
No one starts from zero.
88Each level chooses what it takes
A team decides what it takes from the organisation, and each person what they take from the team.
Shared knowledge helps without overriding local judgement.
89Published versions
What a team or organisation shares is published in versions, and an old version can be retired.
Changes to shared knowledge are deliberate and traceable.
Multimodal intelligence
2Text, documents, images, audio and video, learned in one place.
90Multimodal learning
MS Duplex learns from text, documents, images, audio, video and PDFs, all in one place.
A photo, a report and a call about the same subject are understood together.
91Search across formats
MS Duplex finds the right media by meaning, whatever format the request and the result are in.
A typed question can bring back the right photo, document, audio clip or moment in a video.
Confidence and autonomy
4MS Duplex knows when it knows, and hands over the rest.
92Familiarity recognition
MS Duplex recognises when new work resembles situations it has already learned.
When the same kind of work comes back, what worked last time comes with it.
See it at work →93Uncertainty measurement
MS Duplex weighs how well a decision is supported before anyone relies on it.
You see how sure MS Duplex is, and it steps back when it is not sure enough.
See it at work →94Answering what it knows
MS Duplex handles familiar, well-understood work itself when your model is not needed.
Routine answers come faster, with fewer model calls and lower running costs.
See it at work →95Intelligent escalation
For new or uncertain work, MS Duplex brings in more context, an approved tool, a stronger model or a person.
Deeper reasoning goes where it makes the most difference.
See it at work →
Runtime safety and oversight
9Every action an agent attempts is checked against your controls.
96Behavioural enforcement
MS Duplex checks every command, tool call, file change and outside action an agent attempts.
Safety becomes a check on every action, instead of a request in a prompt.
See it at work →97Human oversight
MS Duplex pauses a sensitive, unusual or uncertain action until a person approves it.
People make the important calls, and routine work runs at full speed.
98Explicit cross-user access
MS Duplex shares private context only when access is granted, and any grant can be withdrawn.
People on the same system see only what has been shared with them.
100Reversible learning
MS Duplex can unlearn a learned label, correction, outcome, workflow, task, thread or example when you ask.
You choose what lasts, and a lesson can be taken back.
See it at work →101Cascading forgetting
When a source is forgotten, MS Duplex also removes the summaries, facts, task handovers and lessons drawn from it.
Deleted information does not live on in a summary or a fact made from it.
See it at work →102Traceable knowledge
MS Duplex links everything it learned to the events, documents, outcomes and sources it came from.
Any conclusion can be followed back to where it started.
See it at work →103Workspace isolation
MS Duplex confines context, learning, tasks, rules, tools and sharing to the workspace they belong to.
What one workspace learns stays inside it, unless you choose to share it.
See it at work →104Organisation-rule precedence
MS Duplex applies mandatory organisation-wide rules ahead of weaker local preferences.
Teams can add stricter controls of their own, and the company’s protections still hold.
See it at work →
Operational intelligence
5Less repeated work, and deeper reasoning where it pays.
105Tool use under your rules
MS Duplex applies the same rules, permissions, confidence and task context to tool execution that it applies to agent decisions.
Before an agent acts outside, MS Duplex checks the action against your rules.
See it at work →106Streaming execution
MS Duplex streams progress and output while an approved tool runs a long job.
Agents stay responsive while the long work runs in the background.
See it at work →107Tool-outcome learning
MS Duplex learns from every tool run, whether it worked or not.
The choice of tool, and how it is used, gets better with every run.
See it at work →108Reduced repeated reasoning
MS Duplex reuses learned context, decisions, skills and outcomes instead of solving the same problem from the beginning.
Answers come faster, with fewer tokens and lower model costs.
See it at work →109Reasoning where it matters
MS Duplex handles familiar work efficiently and directs novel situations towards deeper reasoning.
Routine work stays fast and cheap, and the hard cases get the depth they need.
See it at work →
Integration and portability
5One foundation for every agent, app and model you use.
110Flexible agent integration
MS Duplex works inside the assistants your team already uses, through plugins, and inside the agents your company builds, through MCP + hooks.
Your team gets learning, rules, tasks and tools without changing how it works.
See it at work →111One shared foundation
MS Duplex gives every connected agent the same knowledge, tasks, skills, rules and proven outcomes.
Your separate AI apps work as one system that gets better with use.
See it at work →112Provider-independent intelligence
What MS Duplex learns lives outside any one model, so it outlasts every model you use.
You can move to a better model whenever one arrives, and the knowledge built from your work comes too.
See it at work →113Application-independent intelligence
What MS Duplex learns, the skills it builds and the rules you set work in every supported app.
A new AI app starts with what your work has already taught MS Duplex elsewhere.
See it at work →114An advantage that compounds
MS Duplex turns every completed task into knowledge that can improve the next task.
You rent the model by the call. What MS Duplex learns from your work is an asset you own, and it compounds with every task.
See it at work →
Built for the work your teams do.
The same intelligence in every kind of work, meeting the standard each one sets.
In software engineering, MS Duplex learns your repository’s conventions, and a patch counts once its checks pass.
