<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Huy Nguyen</title>
    <description>The latest articles on DEV Community by Huy Nguyen (@huynguyen294).</description>
    <link>https://dev.to/huynguyen294</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F1040842%2F7ff7476a-b560-4cd4-b82e-c72f90603057.jpeg</url>
      <title>DEV Community: Huy Nguyen</title>
      <link>https://dev.to/huynguyen294</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/huynguyen294"/>
    <language>en</language>
    <item>
      <title>Tutorial 00: AI Artifacts Workflow Guide — When to Use What</title>
      <dc:creator>Huy Nguyen</dc:creator>
      <pubDate>Tue, 17 Nov 2026 16:43:55 +0000</pubDate>
      <link>https://dev.to/huynguyen294/tutorial-00-ai-artifacts-workflow-guide-when-to-use-what-2hjf</link>
      <guid>https://dev.to/huynguyen294/tutorial-00-ai-artifacts-workflow-guide-when-to-use-what-2hjf</guid>
      <description>&lt;p&gt;Not every coding task needs a multi-document workspace or automated multi-agent handoffs. Sometimes you just want to read a 200-line implementation plan in a clean &lt;strong&gt;Preview UI&lt;/strong&gt; instead of squinting at a narrow chat sidebar or raw Markdown syntax. Other times, you want a simple place to collect the 3–4 documents created during a feature, or you need a structured workspace that multiple chats and AI agents can join.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;AI Artifacts&lt;/strong&gt; is designed so that &lt;strong&gt;complexity follows real need&lt;/strong&gt;. Use this guide to choose the exact level of structure your current task requires.&lt;/p&gt;




&lt;h2&gt;
  
  
  Comparison at a Glance
&lt;/h2&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Workflow&lt;/th&gt;
&lt;th&gt;Best For&lt;/th&gt;
&lt;th&gt;What you get&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;
&lt;strong&gt;1. Single-Document Review&lt;/strong&gt;&lt;br&gt;👉 &lt;a href="https://dev.to/huynguyen294/this-extension-lets-your-ai-agent-create-artifacts-for-plans-workarounds-and-designs-in-cursor-19n3"&gt;Part 01&lt;/a&gt;
&lt;/td&gt;
&lt;td&gt;Reviewing one plan, design, or note—or having a 2nd AI review your plan.&lt;/td&gt;
&lt;td&gt;Opens documents in a rendered &lt;strong&gt;Preview UI&lt;/strong&gt; (tables, code, Mermaid) instead of raw Markdown. Highlight any block to leave inline comments, click &lt;strong&gt;Proceed&lt;/strong&gt; to execute, or use &lt;strong&gt;Inspect artifact&lt;/strong&gt; for a 2nd AI check.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;
&lt;strong&gt;2. All Your Session’s Artifacts in One Place&lt;/strong&gt;&lt;br&gt;👉 &lt;a href="https://dev.to/huynguyen294/organizing-work-sessions-from-simple-collections-to-shared-context-hlk"&gt;Part 02&lt;/a&gt;
&lt;/td&gt;
&lt;td&gt;Storing and browsing multiple artifacts created during a single working session.&lt;/td&gt;
&lt;td&gt;Collects your session's artifacts into one visual &lt;strong&gt;Working&lt;/strong&gt; gallery (with the side panel collapsed) featuring 2-line previews, instant title search, and type filters.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;
&lt;strong&gt;3. Work at a Glance &amp;amp; Shared Context&lt;/strong&gt;&lt;br&gt;👉 &lt;a href="https://dev.to/huynguyen294/organizing-work-sessions-from-simple-collections-to-shared-context-hlk"&gt;Part 02&lt;/a&gt; · &lt;a href="https://dev.to/huynguyen294/cross-client-handoffs-reusable-context-across-ai-agents-40d9"&gt;Part 03&lt;/a&gt;
&lt;/td&gt;
&lt;td&gt;Seeing a larger piece of work at a glance and continuing across fresh chats or different AI clients.&lt;/td&gt;
&lt;td&gt;Organizes documents into &lt;strong&gt;Working&lt;/strong&gt;, &lt;strong&gt;References&lt;/strong&gt;, and &lt;strong&gt;Archived&lt;/strong&gt; with shared &lt;strong&gt;Instructions&lt;/strong&gt;. Click &lt;strong&gt;Connect&lt;/strong&gt; so any new chat or AI client picks up the context immediately without re-explaining.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;
&lt;strong&gt;4. Automated Multi-Agent Workflows&lt;/strong&gt;&lt;br&gt;👉 &lt;a href="https://dev.to/huynguyen294/automated-multi-agent-workflows-43mb"&gt;Part 04&lt;/a&gt;
&lt;/td&gt;
&lt;td&gt;Running repeatable multi-agent loops (e.g., &lt;em&gt;Coder ↔ Reviewer&lt;/em&gt;) without copy-pasting between chats.&lt;/td&gt;
&lt;td&gt;Agents follow your &lt;strong&gt;Instructions&lt;/strong&gt; to stand by in the &lt;strong&gt;Waiting room&lt;/strong&gt; and wake each other up—or hand off review feedback through artifacts—while you keep final &lt;strong&gt;Proceed&lt;/strong&gt; control.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;h2&gt;
  
  
  Detailed Breakdown: Which Workflow Fits Your Task?
&lt;/h2&gt;

&lt;h3&gt;
  
  
  1. Single-Document Review
&lt;/h3&gt;

&lt;p&gt;&lt;strong&gt;Why it matters:&lt;/strong&gt; Scrolling through a 200-line plan in a narrow chat sidebar is cramped, and opening a raw &lt;code&gt;.md&lt;/code&gt; file forces you to read unrendered syntax (&lt;code&gt;|---|---|&lt;/code&gt;, raw &lt;code&gt;mermaid&lt;/code&gt; fences, and heading hashes) while manually quoting lines back into chat.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;What you get:&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;A rich reading and commenting UI:&lt;/strong&gt; Every artifact opens in a dedicated &lt;strong&gt;Preview UI&lt;/strong&gt; with rendered tables, syntax-highlighted code, and visual Mermaid diagrams. Highlight any rendered block to attach inline comments or questions, then choose &lt;strong&gt;Review&lt;/strong&gt;, &lt;strong&gt;Proceed&lt;/strong&gt;, or &lt;strong&gt;Just save&lt;/strong&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Easy second-AI review on a single plan:&lt;/strong&gt; Click &lt;strong&gt;&lt;code&gt;Inspect artifact&lt;/code&gt;&lt;/strong&gt; in the preview header and paste the prompt into a second AI agent so it can read the proposal and leave review feedback directly on the artifact.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;👉 &lt;strong&gt;Read the guide:&lt;/strong&gt; &lt;a href="https://dev.to/huynguyen294/this-extension-lets-your-ai-agent-create-artifacts-for-plans-workarounds-and-designs-in-cursor-19n3"&gt;Part 01: Getting Started, Inline Review &amp;amp; Cross-Agent Checks&lt;/a&gt;&lt;/p&gt;

&lt;h3&gt;
  
  
  2. All Your Session’s Artifacts in One Place
&lt;/h3&gt;

&lt;p&gt;&lt;strong&gt;Why it matters:&lt;/strong&gt; During a single coding session, you might generate 3 or 4 artifacts (an investigation note, an API design, a migration checklist). Instead of scrolling up and down a long chat thread to find links, you want one visual place to store, search, and reopen them.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;What you get:&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Automatic collection in one session:&lt;/strong&gt; Ask your AI to start a Work Session for your task—it connects automatically and places every artifact created during the conversation into the session's &lt;strong&gt;&lt;code&gt;Working&lt;/code&gt;&lt;/strong&gt; view.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;A focused visual gallery:&lt;/strong&gt; Collapse the right-hand side panel in the Session Manager for a full-width, multi-column &lt;strong&gt;&lt;code&gt;Working&lt;/code&gt;&lt;/strong&gt; grid showing 2-line previews, instant title search, and type filters (&lt;code&gt;Implementation plan&lt;/code&gt;, &lt;code&gt;Design&lt;/code&gt;, &lt;code&gt;Spec&lt;/code&gt;, &lt;code&gt;Note&lt;/code&gt;, etc.).&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Instant rendered previews:&lt;/strong&gt; Click any card in the gallery to open its &lt;strong&gt;Preview UI&lt;/strong&gt; immediately—no hunting through raw Markdown files.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;👉 &lt;strong&gt;Read the guide:&lt;/strong&gt; &lt;a href="https://dev.to/huynguyen294/organizing-work-sessions-from-simple-collections-to-shared-context-hlk"&gt;Part 02: Organizing Work Sessions — From Simple Collections to Agent Context&lt;/a&gt;&lt;/p&gt;

&lt;h3&gt;
  
  
  3. Work at a Glance &amp;amp; Shared Context
&lt;/h3&gt;

&lt;p&gt;&lt;strong&gt;Why it matters:&lt;/strong&gt; Managing a larger feature through ad-hoc Markdown folders in your repo is messy—you end up reading dry raw &lt;code&gt;.md&lt;/code&gt; files, inventing a new folder structure and workflow prompt every time, cluttering your Git tree, and re-explaining everything when switching chats or AI clients.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;What you get:&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Rich UI to see your work at a glance:&lt;/strong&gt; Instead of a raw Markdown folder, the two-column &lt;strong&gt;Session Manager&lt;/strong&gt; organizes your work into &lt;strong&gt;&lt;code&gt;Working&lt;/code&gt;&lt;/strong&gt; (active drafts), &lt;strong&gt;&lt;code&gt;References&lt;/code&gt;&lt;/strong&gt; (specs and blueprints), and &lt;strong&gt;&lt;code&gt;Archived&lt;/code&gt;&lt;/strong&gt; (completed work). Click any card across any zone to read it in the rendered &lt;strong&gt;Preview UI&lt;/strong&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Built-in structure &amp;amp; AI skill (no DIY setup):&lt;/strong&gt; Write your rules or workflow once in &lt;strong&gt;&lt;code&gt;Instructions&lt;/code&gt;&lt;/strong&gt;. When you click &lt;strong&gt;&lt;code&gt;Connect&lt;/code&gt;&lt;/strong&gt; in a fresh chat or another AI client, the built-in Work Session skill automatically guides the agent to follow your &lt;strong&gt;Instructions&lt;/strong&gt; and read the exact reference documents it needs.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Zero repo clutter:&lt;/strong&gt; All working documents and session context live outside your repository in global storage (with 1-click &lt;strong&gt;Export ZIP&lt;/strong&gt; whenever you want a portable copy).&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;👉 &lt;strong&gt;Read the guides:&lt;/strong&gt; &lt;a href="https://dev.to/huynguyen294/organizing-work-sessions-from-simple-collections-to-shared-context-hlk"&gt;Part 02: Organizing Work Sessions&lt;/a&gt; · &lt;a href="https://dev.to/huynguyen294/cross-client-handoffs-reusable-context-across-ai-agents-40d9"&gt;Part 03: Cross-Client Handoffs &amp;amp; Reusable Context&lt;/a&gt;&lt;/p&gt;

&lt;h3&gt;
  
  
  4. Automated Multi-Agent Workflows
&lt;/h3&gt;

&lt;p&gt;&lt;strong&gt;Why it matters:&lt;/strong&gt; Acting as a manual messenger between two agents in the same session (&lt;em&gt;"Agent A finished updating the plan, now tell Agent B to review it"&lt;/em&gt;) gets repetitive when you run structured review loops.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;What you get:&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Role-based coordination:&lt;/strong&gt; Define roles and handoff rules in &lt;strong&gt;Session Instructions&lt;/strong&gt; (for example, &lt;em&gt;Coder&lt;/em&gt;, &lt;em&gt;Reviewer&lt;/em&gt;, and &lt;em&gt;Assessor&lt;/em&gt;).&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Automated handoffs:&lt;/strong&gt; Participating agents stand by in the &lt;strong&gt;Waiting room&lt;/strong&gt; and wake each other up automatically when work is ready—or hand review feedback directly from one agent to another through the artifact.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Human control where it counts:&lt;/strong&gt; Monitor active waits in &lt;strong&gt;&lt;code&gt;Waiting room (n)&lt;/code&gt;&lt;/strong&gt;, stop any wait with one click, and keep final authority over &lt;strong&gt;Review&lt;/strong&gt; and &lt;strong&gt;Proceed&lt;/strong&gt;.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;👉 &lt;strong&gt;Read the guide:&lt;/strong&gt; &lt;a href="https://dev.to/huynguyen294/automated-multi-agent-workflows-43mb"&gt;Part 04: Automated Multi-Agent Workflows with Session Waiters &amp;amp; Waiting Room&lt;/a&gt;&lt;/p&gt;

</description>
      <category>vscode</category>
      <category>ai</category>
      <category>mpc</category>
      <category>agents</category>
    </item>
    <item>
      <title>Automated Multi-Agent Workflows</title>
      <dc:creator>Huy Nguyen</dc:creator>
      <pubDate>Tue, 17 Nov 2026 16:43:25 +0000</pubDate>
      <link>https://dev.to/huynguyen294/automated-multi-agent-workflows-43mb</link>
      <guid>https://dev.to/huynguyen294/automated-multi-agent-workflows-43mb</guid>
      <description>&lt;p&gt;When you run a repeatable multi-agent process—such as one agent drafting or implementing a plan, a second agent reviewing the work for issues, and a third step assessing which findings are actually worth fixing—manually relaying messages between chats (&lt;em&gt;"Hey Reviewer, the author just finished updating the artifact, please check it now"&lt;/em&gt;) quickly becomes tedious.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;AI Artifacts&lt;/strong&gt; lets participating agents coordinate handoffs inside a shared &lt;strong&gt;Work Session&lt;/strong&gt; without you copying and pasting between chats. You define the roles, handoff rules, and approval checkpoints in &lt;strong&gt;Session Instructions&lt;/strong&gt;, monitor standing waits in the &lt;strong&gt;Waiting room&lt;/strong&gt;, and step in only where human judgment matters.&lt;/p&gt;




&lt;h2&gt;
  
  
  How Automated Coordination Works (Without Giving Up Control)
&lt;/h2&gt;

&lt;p&gt;Automation in AI Artifacts is &lt;strong&gt;instruction-driven, not a black-box pipeline&lt;/strong&gt;. The extension provides the shared &lt;strong&gt;Work Session&lt;/strong&gt;, the rendered &lt;strong&gt;Preview UI&lt;/strong&gt;, and the handoff mechanisms; &lt;strong&gt;you&lt;/strong&gt; choose the roles, which AI clients to use, and where agents must stop for your decision.&lt;/p&gt;

&lt;p&gt;Inside a Work Session, agents can hand work over in two ways:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Agent-to-Agent Review Through an Artifact:&lt;/strong&gt; One agent waits for feedback on a specific artifact while a &lt;strong&gt;Reviewer&lt;/strong&gt; agent inspects the document, posts findings as artifact comments, and submits &lt;strong&gt;Review&lt;/strong&gt;. The waiting agent receives the feedback automatically and continues according to your &lt;strong&gt;Instructions&lt;/strong&gt; (for example, preparing an assessment of the findings for your review).&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Standby Handoffs via the &lt;code&gt;Waiting room&lt;/code&gt;:&lt;/strong&gt; An agent can stand by in the session's &lt;strong&gt;Waiting room&lt;/strong&gt; for its next assignment. When another agent finishes its step, it wakes that specific waiting agent with a handoff message and target artifact.&lt;/li&gt;
&lt;/ol&gt;

&lt;blockquote&gt;
&lt;p&gt;&lt;strong&gt;You always keep final authority:&lt;/strong&gt; Joining a session, reading an artifact, or receiving a handoff never authorizes an agent to execute an unapproved plan. Only your explicit click on &lt;strong&gt;&lt;code&gt;Proceed&lt;/code&gt;&lt;/strong&gt; on a plan artifact authorizes code execution.&lt;/p&gt;
&lt;/blockquote&gt;




&lt;h2&gt;
  
  
  Step 1: Define Roles and Checkpoints in Session &lt;code&gt;Instructions&lt;/code&gt;
&lt;/h2&gt;

&lt;p&gt;Open your &lt;strong&gt;Work Session Manager&lt;/strong&gt;, expand the right-hand side panel, and define how your agents should collaborate in the &lt;strong&gt;Instructions&lt;/strong&gt; block (or ask your connected agent in chat to help draft and save the rules for you).&lt;/p&gt;

&lt;p&gt;Because the built-in skills already know how to use the underlying tools, you can write your &lt;strong&gt;Instructions&lt;/strong&gt; in plain language:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight markdown"&gt;&lt;code&gt;&lt;span class="gu"&gt;# Session Workflow Instructions
&lt;/span&gt;
&lt;span class="gu"&gt;## Roles
&lt;/span&gt;&lt;span class="k"&gt;- &lt;/span&gt;&lt;span class="gs"&gt;**Author / Implementer:**&lt;/span&gt; Drafts plans and documents in &lt;span class="sb"&gt;`Working`&lt;/span&gt;, evaluates review findings, and implements code changes ONLY after I click &lt;span class="sb"&gt;`Proceed`&lt;/span&gt;.
&lt;span class="k"&gt;- &lt;/span&gt;&lt;span class="gs"&gt;**Reviewer:**&lt;/span&gt; Checks the target artifact against the requirements and design in &lt;span class="sb"&gt;`References`&lt;/span&gt;, posts each finding as a comment on the artifact, and submits Review without modifying the document directly.

&lt;span class="gu"&gt;## Handoff &amp;amp; Assessment Rules
&lt;/span&gt;1. When reviewing a draft artifact, the Author waits for review on that artifact while the Reviewer inspects it and submits Review.
2. When the Author receives the Reviewer's feedback, do not apply every suggestion blindly. Create a &lt;span class="sb"&gt;`Review Assessment`&lt;/span&gt; artifact in &lt;span class="sb"&gt;`Working`&lt;/span&gt; that keeps the original findings alongside an assessment of each item's validity, necessity, scope, and risk.
3. Stop and report to me so I can review the assessment in the Preview UI and decide which changes to apply.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Click &lt;strong&gt;Save&lt;/strong&gt; in the Instructions panel (or edit &lt;code&gt;instructions.md&lt;/code&gt; via &lt;strong&gt;Open Markdown File&lt;/strong&gt;).&lt;/p&gt;




&lt;h2&gt;
  
  
  Step 2: Run an Agent-to-Agent Review Handoff
&lt;/h2&gt;

&lt;p&gt;With your workflow saved in &lt;strong&gt;Instructions&lt;/strong&gt;, connect two AI chats (across the same editor or two different clients like &lt;strong&gt;Cursor&lt;/strong&gt; and &lt;strong&gt;Claude Code&lt;/strong&gt; / &lt;strong&gt;Codex&lt;/strong&gt;) to the same Work Session using &lt;strong&gt;&lt;code&gt;Connect&lt;/code&gt;&lt;/strong&gt;:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Assign the Author / Supporter role in Chat 1:&lt;/strong&gt;
&lt;/li&gt;
&lt;/ol&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;   [Pasted Session Connect Prompt]
   You are the Author. Wait for review submission on the current artifact in Working. When you receive the Reviewer's feedback, create the Review Assessment artifact according to our Session Instructions and wait for my decision.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Assign the Reviewer role in Chat 2:&lt;/strong&gt;
&lt;/li&gt;
&lt;/ol&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;   [Pasted Session Connect Prompt]
   You are the Reviewer. Review the active artifact in Working against our Session Context and References. Post each finding as a comment on that artifact, submit Review, and stop without editing the document.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Frc0zur38j2fk9ms1g8k4.gif" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Frc0zur38j2fk9ms1g8k4.gif" alt="The Reviewer submits artifact comments and the waiting agent receives the feedback" width="720" height="404"&gt;&lt;/a&gt;&lt;br&gt;
&lt;em&gt;One agent posts comments and submits Review on the artifact; the waiting agent receives the feedback automatically without copying between chats.&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;What happens next:&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;The &lt;strong&gt;Reviewer&lt;/strong&gt; inspects the document, attaches its comments, and submits &lt;strong&gt;Review&lt;/strong&gt;.&lt;/li&gt;
&lt;li&gt;The &lt;strong&gt;Author&lt;/strong&gt; receives the submitted comments immediately, refreshes the source artifact, and creates the &lt;strong&gt;Review Assessment&lt;/strong&gt; artifact in &lt;strong&gt;&lt;code&gt;Working&lt;/code&gt;&lt;/strong&gt;.&lt;/li&gt;
&lt;li&gt;You open the &lt;strong&gt;Review Assessment&lt;/strong&gt; card in the &lt;strong&gt;Preview UI&lt;/strong&gt; to read the Reviewer's original findings side by side with the Author's assessment—so you can approve only the changes that actually make sense.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fpa2uw4zl5x8i3tfkdfb3.gif" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fpa2uw4zl5x8i3tfkdfb3.gif" alt="The waiting agent creates a review assessment artifact in the Work Session" width="720" height="404"&gt;&lt;/a&gt;&lt;br&gt;
&lt;em&gt;Read the original findings and the assessment together in the Preview UI before deciding what to change.&lt;/em&gt;&lt;/p&gt;




&lt;h2&gt;
  
  
  Step 3: Keep Agents on Standby in the &lt;code&gt;Waiting room&lt;/code&gt;
&lt;/h2&gt;

&lt;p&gt;When you want an agent to stay on standby across multiple tasks in a session—so another agent can wake it up whenever new work is ready—use the &lt;strong&gt;Waiting room&lt;/strong&gt;:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Ask an agent to wait in the session:&lt;/strong&gt;
&lt;/li&gt;
&lt;/ol&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;   Wait in this session for the next review handoff.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Check &lt;code&gt;Waiting room (n)&lt;/code&gt; in the Session Manager:&lt;/strong&gt; Look next to the &lt;strong&gt;&lt;code&gt;＋&lt;/code&gt; (Add Artifact)&lt;/strong&gt; button at the top of the &lt;strong&gt;&lt;code&gt;Working&lt;/code&gt;&lt;/strong&gt; column. Clicking &lt;strong&gt;&lt;code&gt;Waiting room (1)&lt;/code&gt;&lt;/strong&gt; shows every agent currently on standby, what it is waiting for, and the related artifact title on hover.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Wake the waiting agent from another chat:&lt;/strong&gt; When the other agent finishes preparing an artifact or completing a step, ask it (or configure your &lt;strong&gt;Instructions&lt;/strong&gt; so it knows) to send a handoff message to the waiting agent in the session. Only the targeted waiting agent is awakened to run its task.&lt;/li&gt;
&lt;/ol&gt;

&lt;h3&gt;
  
  
  Stay in control at all times
&lt;/h3&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;One-click &lt;code&gt;Stop&lt;/code&gt;:&lt;/strong&gt; Every entry in &lt;strong&gt;&lt;code&gt;Waiting room (n)&lt;/code&gt;&lt;/strong&gt; has a &lt;strong&gt;&lt;code&gt;Stop&lt;/code&gt;&lt;/strong&gt; button. Clicking &lt;strong&gt;&lt;code&gt;Stop&lt;/code&gt;&lt;/strong&gt; ends that wait cleanly without deleting any artifacts or session data, and a stopped wait never restarts on its own.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Works even with the manager tab closed:&lt;/strong&gt; Once agents are connected, background handoffs continue working even if you close the Work Session Manager tab.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Pause auto-open when agents are busy:&lt;/strong&gt; If background rounds open tabs while you are reading something else, open the &lt;strong&gt;Settings (gear)&lt;/strong&gt; menu in the Session Manager header and choose &lt;strong&gt;Turn off auto-open artifact&lt;/strong&gt;. You can still open any card manually in the &lt;strong&gt;Preview UI&lt;/strong&gt; whenever you are ready.&lt;/li&gt;
&lt;/ul&gt;




&lt;h2&gt;
  
  
  Series Summary: Choose the Workflow That Fits Your Task
&lt;/h2&gt;

&lt;p&gt;You now have the complete &lt;strong&gt;AI Artifacts&lt;/strong&gt; toolkit—from reviewing a single document to running coordinated multi-agent sessions:&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Workflow&lt;/th&gt;
&lt;th&gt;Best For&lt;/th&gt;
&lt;th&gt;Guide&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;1. Single-Document Review&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;Reviewing one plan, design, or note in the rendered &lt;strong&gt;Preview UI&lt;/strong&gt; with inline comments—or having a 2nd AI review your plan via &lt;strong&gt;&lt;code&gt;Inspect artifact&lt;/code&gt;&lt;/strong&gt;.&lt;/td&gt;
&lt;td&gt;
&lt;a href="https://dev.to/huynguyen294/tutorial-00-ai-artifacts-workflow-guide-when-to-use-what-2hjf"&gt;Workflow Overview&lt;/a&gt; · &lt;a href="https://dev.to/huynguyen294/this-extension-lets-your-ai-agent-create-artifacts-for-plans-workarounds-and-designs-in-cursor-19n3"&gt;Part 01&lt;/a&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;2. All Your Session’s Artifacts in One Place&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;Storing, searching, and filtering multiple artifacts from one working session in a clean &lt;strong&gt;&lt;code&gt;Working&lt;/code&gt;&lt;/strong&gt; gallery (with the side panel collapsed).&lt;/td&gt;
&lt;td&gt;&lt;a href="https://dev.to/huynguyen294/organizing-work-sessions-from-simple-collections-to-shared-context-hlk"&gt;Part 02&lt;/a&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;3. Work at a Glance &amp;amp; Shared Context&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;Seeing a larger feature at a glance across &lt;strong&gt;&lt;code&gt;Working&lt;/code&gt; / &lt;code&gt;References&lt;/code&gt; / &lt;code&gt;Archived&lt;/code&gt;&lt;/strong&gt; and continuing across fresh chats or different AI clients via &lt;strong&gt;&lt;code&gt;Instructions&lt;/code&gt;&lt;/strong&gt; and &lt;strong&gt;&lt;code&gt;Connect&lt;/code&gt;&lt;/strong&gt;.&lt;/td&gt;
&lt;td&gt;
&lt;a href="https://dev.to/huynguyen294/organizing-work-sessions-from-simple-collections-to-shared-context-hlk"&gt;Part 02&lt;/a&gt; · &lt;a href="https://dev.to/huynguyen294/cross-client-handoffs-reusable-context-across-ai-agents-40d9"&gt;Part 03&lt;/a&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;4. Automated Multi-Agent Workflows&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;Letting connected agents hand off reviews and wake each other up through &lt;strong&gt;&lt;code&gt;Instructions&lt;/code&gt;&lt;/strong&gt; and the &lt;strong&gt;&lt;code&gt;Waiting room&lt;/code&gt;&lt;/strong&gt; while you keep final &lt;strong&gt;&lt;code&gt;Proceed&lt;/code&gt;&lt;/strong&gt; control.&lt;/td&gt;
&lt;td&gt;&lt;strong&gt;You are here (Part 04)&lt;/strong&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

</description>
      <category>ai</category>
      <category>agents</category>
      <category>mcp</category>
      <category>vscode</category>
    </item>
    <item>
      <title>Cross-Client Handoffs &amp; Reusable Context Across AI Agents</title>
      <dc:creator>Huy Nguyen</dc:creator>
      <pubDate>Tue, 17 Nov 2026 16:40:43 +0000</pubDate>
      <link>https://dev.to/huynguyen294/cross-client-handoffs-reusable-context-across-ai-agents-40d9</link>
      <guid>https://dev.to/huynguyen294/cross-client-handoffs-reusable-context-across-ai-agents-40d9</guid>
      <description>&lt;h1&gt;
  
  
  Cross-Client Handoffs &amp;amp; Reusable Context Across AI Agents
&lt;/h1&gt;

&lt;p&gt;Different AI coding clients and models have different strengths. You might prefer &lt;strong&gt;Claude Code&lt;/strong&gt; or &lt;strong&gt;Codex&lt;/strong&gt; for deep architectural analysis and bug investigation, &lt;strong&gt;Cursor&lt;/strong&gt; or &lt;strong&gt;Windsurf&lt;/strong&gt; for multi-file implementation, and &lt;strong&gt;GitHub Copilot&lt;/strong&gt;, &lt;strong&gt;Cline&lt;/strong&gt;, or &lt;strong&gt;Antigravity&lt;/strong&gt; for targeted reviews.&lt;/p&gt;

&lt;p&gt;Normally, switching tools mid-feature—or even opening a fresh chat in the same editor to avoid context degradation ("context rot")—means copying long chat transcripts or explaining your architecture all over again. &lt;strong&gt;AI Artifacts&lt;/strong&gt; removes that friction: because &lt;strong&gt;Preview UI&lt;/strong&gt; rendering, artifact lifecycles, and &lt;strong&gt;Work Sessions&lt;/strong&gt; are shared across all 7 supported AI clients, you can hand work off between chats and agents cleanly without losing momentum.&lt;/p&gt;




&lt;h2&gt;
  
  
  Why Synthesized Context Beats Copying Chat History
&lt;/h2&gt;

&lt;p&gt;A &lt;strong&gt;Work Session&lt;/strong&gt; does &lt;strong&gt;not&lt;/strong&gt; transfer raw chat transcripts from one AI client to another—and that is intentional. A long chat thread is full of dead ends, rejected drafts, and back-and-forth troubleshooting that confuses the next model.&lt;/p&gt;

&lt;p&gt;Instead, reliable handoffs come from two things you control in your Work Session:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Distilled documents&lt;/strong&gt; organized into &lt;strong&gt;&lt;code&gt;Working&lt;/code&gt;&lt;/strong&gt; (active tasks), &lt;strong&gt;&lt;code&gt;References&lt;/code&gt;&lt;/strong&gt; (agreed specs, blueprints, and session context), and &lt;strong&gt;&lt;code&gt;Archived&lt;/code&gt;&lt;/strong&gt; (finished phases).&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Shared &lt;code&gt;Instructions&lt;/code&gt;&lt;/strong&gt; that tell any incoming agent how the session is organized, what role to play, and where to stop for your review.&lt;/li&gt;
&lt;/ul&gt;




&lt;h2&gt;
  
  
  Step 1: Leave a Clean Checkpoint Before Switching Chats or Clients
&lt;/h2&gt;

&lt;p&gt;Before ending a long analytical chat or switching from one AI client to another, ask your current agent to capture the conclusions in a context artifact (and place it in &lt;strong&gt;&lt;code&gt;References&lt;/code&gt;&lt;/strong&gt;):&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;Update this session's context artifact with what we decided, why we chose this approach, what remains open, and the exact next step for a new chat.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;&lt;strong&gt;Why this habit matters:&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Decisions stay explicit:&lt;/strong&gt; Marking what is settled versus what is still open prevents the next agent from treating a rough idea as an approved requirement.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Instant orientation:&lt;/strong&gt; When a new agent joins the session later, it can read your context artifact and immediately understand what is finished (in &lt;code&gt;Archived&lt;/code&gt;), what is in progress (in &lt;code&gt;Working&lt;/code&gt;), and what needs to happen next.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Reusable knowledge:&lt;/strong&gt; If you solved a tricky bug or edge case, capturing the final takeaway as a note or spec in &lt;code&gt;References&lt;/code&gt; keeps that knowledge available for future tasks without cluttering your Git repository.&lt;/li&gt;
&lt;/ul&gt;




&lt;h2&gt;
  
  
  Step 2: Shape &lt;code&gt;Instructions&lt;/code&gt; Together with Your Agent
&lt;/h2&gt;

&lt;p&gt;You do not need to write a complex workflow specification from scratch before starting. Describe how you want to work in plain language and ask your connected agent to help formulate the rules for the session:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;Help me define the Instructions for this session: before proposing code changes, agents should read the Architecture Design and Session Context in References, draft an implementation-plan artifact for my review, and wait for my approval before modifying source files. Save the Instructions once we agree.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;You can also edit the &lt;strong&gt;Instructions&lt;/strong&gt; box directly in the &lt;strong&gt;Work Session Manager&lt;/strong&gt; (or click &lt;strong&gt;Open Markdown File&lt;/strong&gt; to edit &lt;code&gt;instructions.md&lt;/code&gt; in VS Code) and click &lt;strong&gt;Save&lt;/strong&gt;.&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;&lt;strong&gt;Tip:&lt;/strong&gt; If you update &lt;strong&gt;Instructions&lt;/strong&gt; while other chats are already working in the session, simply ask those existing chats to reload the session's Instructions so they pick up the latest rules.&lt;/p&gt;
&lt;/blockquote&gt;




&lt;h2&gt;
  
  
  Step 3: Connect a Fresh Chat or Switch to Another AI Client
&lt;/h2&gt;

&lt;p&gt;Now that your conclusions are saved in artifacts and your rules are in &lt;strong&gt;Instructions&lt;/strong&gt;, switching to a fresh chat—or handing implementation over to another AI client—takes one step:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;In the &lt;strong&gt;Work Session Manager&lt;/strong&gt; header (or from the &lt;strong&gt;Sessions&lt;/strong&gt; sidebar), click &lt;strong&gt;&lt;code&gt;Connect&lt;/code&gt;&lt;/strong&gt; to copy the session join prompt.&lt;/li&gt;
&lt;li&gt;Open a new chat in your target AI client (for example, switching from &lt;strong&gt;Claude Code&lt;/strong&gt; to &lt;strong&gt;Cursor&lt;/strong&gt; or &lt;strong&gt;GitHub Copilot&lt;/strong&gt;) and paste the prompt.&lt;/li&gt;
&lt;li&gt;Ask the new agent to confirm the current state or start the next task:
&lt;/li&gt;
&lt;/ol&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;[Pasted Session Connect Prompt]
Read the Session Context and Instructions. Summarize where we are, then create an implementation-plan artifact for Phase 2 so I can review it.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2F6tbaphe6tw334z2w3oa7.gif" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2F6tbaphe6tw334z2w3oa7.gif" alt="Two agents join the same Work Session and report its progress and next steps" width="759" height="426"&gt;&lt;/a&gt;&lt;br&gt;
&lt;em&gt;New agents join the same Work Session, read the shared context, and pick up right where the previous chat left off.&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;What happens next:&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;The built-in Work Session skill loads your &lt;strong&gt;Instructions&lt;/strong&gt; and the categorized catalog (&lt;code&gt;Working&lt;/code&gt;, &lt;code&gt;References&lt;/code&gt;, &lt;code&gt;Archived&lt;/code&gt;).&lt;/li&gt;
&lt;li&gt;The agent reads the relevant reference documents on demand—without you copying file paths or re-explaining your architecture.&lt;/li&gt;
&lt;li&gt;It creates the &lt;em&gt;Phase 2 Implementation Plan&lt;/em&gt; inside &lt;strong&gt;&lt;code&gt;Working&lt;/code&gt;&lt;/strong&gt; and opens the rendered &lt;strong&gt;Preview UI&lt;/strong&gt; (tables, code, Mermaid) for your inline review.&lt;/li&gt;
&lt;li&gt;Once you refine the plan and click &lt;strong&gt;Proceed&lt;/strong&gt;, the agent executes the approved changes immediately. When the phase is done, ask the agent to move the completed plan into &lt;strong&gt;&lt;code&gt;Archived&lt;/code&gt;&lt;/strong&gt; to keep &lt;strong&gt;&lt;code&gt;Working&lt;/code&gt;&lt;/strong&gt; focused.&lt;/li&gt;
&lt;/ul&gt;




&lt;h2&gt;
  
  
  Step 4: Bring in a Reviewer Agent — and Assess Findings Before Fixing
&lt;/h2&gt;

&lt;p&gt;When working on an important feature, you can connect a second AI client (for example, &lt;strong&gt;Codex&lt;/strong&gt; or &lt;strong&gt;Claude&lt;/strong&gt;) to the same Work Session as an independent &lt;strong&gt;Reviewer&lt;/strong&gt;—and use an &lt;strong&gt;Assessment&lt;/strong&gt; step so you never waste time on noisy or unnecessary suggestions:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Have the Reviewer inspect the work:&lt;/strong&gt; Click &lt;strong&gt;&lt;code&gt;Connect&lt;/code&gt;&lt;/strong&gt;, paste the prompt into your Reviewer chat, and ask it to check the active artifact in &lt;code&gt;Working&lt;/code&gt; against the requirements in &lt;code&gt;References&lt;/code&gt;. The Reviewer can post its findings directly as comments on the artifact (or create a findings artifact in &lt;code&gt;Working&lt;/code&gt;).&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Assess findings before choosing fixes:&lt;/strong&gt; Instead of blindly applying every suggestion—or manually reading through dozens of minor nitpicks—ask your primary agent (or a dedicated Assessor agent) to evaluate the findings first:
&lt;/li&gt;
&lt;/ol&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;Assess these review findings for validity, necessity, scope, and regression risk. Keep the original findings visible alongside your assessment in a review artifact and wait for my decision before applying any changes.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2F7ljmsqj5wcrrjsbytkc6.gif" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2F7ljmsqj5wcrrjsbytkc6.gif" alt=" " width="720" height="404"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Decide in the Preview UI:&lt;/strong&gt; Open the assessment artifact in the &lt;strong&gt;Preview UI&lt;/strong&gt; to read the original findings side by side with the assessment. Use inline comments to confirm which fixes you actually want, then let the agent apply only the approved changes.&lt;/li&gt;
&lt;/ol&gt;




&lt;h2&gt;
  
  
  What's next?
&lt;/h2&gt;

&lt;p&gt;In this guide, you stayed in the driver's seat between phases—clicking &lt;strong&gt;&lt;code&gt;Connect&lt;/code&gt;&lt;/strong&gt; when opening a new chat or bringing in a Reviewer. What if you have two or three agents open in the same Work Session and want them to &lt;strong&gt;hand work off to each other automatically&lt;/strong&gt; while preserving your review checkpoints?&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Workflow overview:&lt;/strong&gt; &lt;a href="https://dev.to/huynguyen294/tutorial-00-ai-artifacts-workflow-guide-when-to-use-what-2hjf"&gt;Part 00: AI Artifacts Workflow Guide — When to Use What&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Previous guide:&lt;/strong&gt; &lt;a href="https://dev.to/huynguyen294/organizing-work-sessions-from-simple-collections-to-shared-context-hlk"&gt;Part 02: Organizing Work Sessions — From Simple Collections to Shared Context&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Next guide:&lt;/strong&gt; &lt;a href="https://dev.to/huynguyen294/automated-multi-agent-workflows-43mb"&gt;Part 04: Automated Multi-Agent Workflows with Session Waiters &amp;amp; Waiting Room&lt;/a&gt;
&lt;/li&gt;
&lt;/ul&gt;

</description>
      <category>ai</category>
      <category>agents</category>
      <category>mcp</category>
      <category>vscode</category>
    </item>
    <item>
      <title>Organizing Work Sessions — From Simple Collections to Shared Context</title>
      <dc:creator>Huy Nguyen</dc:creator>
      <pubDate>Tue, 17 Nov 2026 16:35:55 +0000</pubDate>
      <link>https://dev.to/huynguyen294/organizing-work-sessions-from-simple-collections-to-shared-context-hlk</link>
      <guid>https://dev.to/huynguyen294/organizing-work-sessions-from-simple-collections-to-shared-context-hlk</guid>
      <description>&lt;p&gt;Sometimes a single artifact is not enough. Even in one coding session, you might generate an investigation note, an API design, and an implementation plan. And when a feature stretches across multiple days, fresh chats, or different AI clients, keeping track of which documents are active, which are reference specs, and which are finished becomes a chore.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Work Sessions&lt;/strong&gt; scale naturally with your task so you only use as much structure as you need:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;All Your Session’s Artifacts in One Place:&lt;/strong&gt; Start a session and &lt;strong&gt;collapse the side panel&lt;/strong&gt; for a clean, full-width &lt;strong&gt;&lt;code&gt;Working&lt;/code&gt;&lt;/strong&gt; gallery with 2-line previews, instant title search, and type filters—where every card opens directly in the rendered &lt;strong&gt;Preview UI&lt;/strong&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Work at a Glance &amp;amp; Shared Context:&lt;/strong&gt; Expand the &lt;strong&gt;two-column Session Manager&lt;/strong&gt; to organize documents across &lt;code&gt;Working&lt;/code&gt;, &lt;code&gt;References&lt;/code&gt;, and &lt;code&gt;Archived&lt;/code&gt; zones, write shared &lt;strong&gt;&lt;code&gt;Instructions&lt;/code&gt;&lt;/strong&gt;, and click &lt;strong&gt;&lt;code&gt;Connect&lt;/code&gt;&lt;/strong&gt; so any new chat or AI client picks up your project context immediately—without cluttering your Git repository with ad-hoc Markdown folders.&lt;/li&gt;
&lt;/ol&gt;




&lt;h2&gt;
  
  
  1. All Your Session’s Artifacts in One Place
&lt;/h2&gt;

&lt;p&gt;If you just want one visual place to store, search, and reopen all the artifacts created during a working session—without setting up multi-zone rules or agent instructions—a Work Session takes seconds to use.&lt;/p&gt;

&lt;h3&gt;
  
  
  Start a session (from chat or from an open artifact)
&lt;/h3&gt;

&lt;p&gt;You can start a Work Session in two ways:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Ask your AI directly in chat:&lt;/strong&gt; Ask the agent to start a new Work Session for your task:
&lt;/li&gt;
&lt;/ul&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;  Start a Work Session named "Auth Token Refresh" and create an API design artifact for me to review.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;The agent creates the session, connects automatically, and places every artifact created during that conversation into the session's &lt;strong&gt;&lt;code&gt;Working&lt;/code&gt;&lt;/strong&gt; view.&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;From an existing artifact in the IDE:&lt;/strong&gt; While viewing an artifact in the &lt;strong&gt;Preview UI&lt;/strong&gt;, click &lt;strong&gt;&lt;code&gt;Start a work session&lt;/code&gt;&lt;/strong&gt; in the top header (or click &lt;strong&gt;&lt;code&gt;+&lt;/code&gt;&lt;/strong&gt; in the &lt;strong&gt;Sessions&lt;/strong&gt; section of the AI Artifacts sidebar), enter a title, and paste the copied connect prompt into your chat so future artifacts in that thread are added automatically.&lt;/li&gt;
&lt;/ul&gt;

&lt;h3&gt;
  
  
  Collapse the side panel for a focused &lt;code&gt;Working&lt;/code&gt; gallery
&lt;/h3&gt;

&lt;p&gt;If you do not need &lt;code&gt;Instructions&lt;/code&gt;, &lt;code&gt;References&lt;/code&gt;, or &lt;code&gt;Archived&lt;/code&gt; zones yet, click the &lt;strong&gt;Side Panel Toggle&lt;/strong&gt; button in the Session Manager header to collapse the right-hand panel.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Frtaltmzkdvg29vpv42vn.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Frtaltmzkdvg29vpv42vn.png" alt="Work Session Manager with the side panel collapsed showing the Working grid" width="800" height="431"&gt;&lt;/a&gt;&lt;br&gt;
&lt;em&gt;Collapse the side panel to turn the Work Session Manager into a focused, multi-column Working gallery.&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;With the side panel hidden, the Session Manager acts as a clean visual dashboard for your session:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Responsive Card Grid:&lt;/strong&gt; Your &lt;strong&gt;&lt;code&gt;Working&lt;/code&gt;&lt;/strong&gt; artifacts expand across up to 3 columns, showing each document's title, type badge, 2-line preview snippet, and date.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Instant Search &amp;amp; Type Filtering:&lt;/strong&gt; Use the &lt;strong&gt;Filter&lt;/strong&gt; input at the top to search cards by title, or use the &lt;strong&gt;Type Filter&lt;/strong&gt; dropdown (&lt;code&gt;Implementation plan&lt;/code&gt;, &lt;code&gt;Design&lt;/code&gt;, &lt;code&gt;Spec&lt;/code&gt;, &lt;code&gt;Note&lt;/code&gt;, etc.) to narrow down documents immediately.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Add Existing Artifacts Anytime:&lt;/strong&gt; Click the &lt;strong&gt;&lt;code&gt;＋&lt;/code&gt; (Add Artifact)&lt;/strong&gt; button beside &lt;code&gt;Working&lt;/code&gt; to pull in any standalone artifact from your global library.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;One-Click Preview UI:&lt;/strong&gt; Click any card to open its rendered &lt;strong&gt;Preview UI&lt;/strong&gt; with formatted tables, Mermaid diagrams, Shiki code blocks, and inline comments—no hunting through raw Markdown files.&lt;/li&gt;
&lt;/ul&gt;




&lt;h2&gt;
  
  
  2. Work at a Glance &amp;amp; Shared Context
&lt;/h2&gt;

&lt;p&gt;Managing a larger feature through ad-hoc Markdown folders inside your repository gets messy fast—you end up reading dry raw &lt;code&gt;.md&lt;/code&gt; files, inventing a different folder convention and prompt template every time, polluting &lt;code&gt;git status&lt;/code&gt;, and re-explaining the project whenever you open a fresh chat.&lt;/p&gt;

&lt;p&gt;When your work grows beyond a few files, click the &lt;strong&gt;Side Panel Toggle&lt;/strong&gt; to expand the right-hand panel. Now the &lt;strong&gt;Work Session&lt;/strong&gt; gives both &lt;strong&gt;you&lt;/strong&gt; and &lt;strong&gt;your AI agents&lt;/strong&gt; a complete, ready-made environment stored outside your Git repository:&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Who It Serves&lt;/th&gt;
&lt;th&gt;What You Get&lt;/th&gt;
&lt;th&gt;Why It Matters&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;You (In the IDE)&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;
&lt;strong&gt;Two-Column Session Manager &amp;amp; Preview UI:&lt;/strong&gt; &lt;code&gt;Working&lt;/code&gt; cards on the left; &lt;code&gt;Session Type&lt;/code&gt;, &lt;code&gt;Instructions&lt;/code&gt;, &lt;code&gt;References&lt;/code&gt;, and &lt;code&gt;Archived&lt;/code&gt; on the right. Every card across all zones opens in the rendered &lt;strong&gt;Preview UI&lt;/strong&gt;.&lt;/td&gt;
&lt;td&gt;You see the entire state of a feature &lt;strong&gt;at a glance&lt;/strong&gt; and read formatted previews instead of managing raw &lt;code&gt;.md&lt;/code&gt; files in your repo.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;Your AI Agents&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;
&lt;strong&gt;Built-In Skill &amp;amp; Prepared Context (&lt;code&gt;Connect&lt;/code&gt;):&lt;/strong&gt; The bundled Work Session skill reads your Markdown &lt;code&gt;Instructions&lt;/code&gt; and the categorized catalog of &lt;code&gt;Working&lt;/code&gt;, &lt;code&gt;References&lt;/code&gt;, and &lt;code&gt;Archived&lt;/code&gt; documents.&lt;/td&gt;
&lt;td&gt;Any new chat or supported AI client knows your rules and reads the exact reference documents it needs without you inventing folder conventions or pasting file paths.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;h3&gt;
  
  
  Organize documents across &lt;code&gt;Working&lt;/code&gt;, &lt;code&gt;References&lt;/code&gt;, and &lt;code&gt;Archived&lt;/code&gt;
&lt;/h3&gt;

&lt;p&gt;Use the card menu in the Session Manager—or simply ask your connected AI in chat—to move and reorder documents across the three zones as your feature evolves:&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Zone&lt;/th&gt;
&lt;th&gt;Location in UI&lt;/th&gt;
&lt;th&gt;Purpose for You&lt;/th&gt;
&lt;th&gt;How Agents Use It&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;Working&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;Main left column&lt;/td&gt;
&lt;td&gt;Active documents currently being drafted, reviewed, or implemented (e.g., &lt;em&gt;Phase 2 Implementation Plan&lt;/em&gt;).&lt;/td&gt;
&lt;td&gt;Treated as the current work in progress.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;References&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;Right side panel&lt;/td&gt;
&lt;td&gt;Stable, approved context that governs the work (e.g., &lt;em&gt;Architecture Blueprint&lt;/em&gt;, &lt;em&gt;API Spec&lt;/em&gt;, &lt;em&gt;Requirements&lt;/em&gt;).&lt;/td&gt;
&lt;td&gt;Consulted on demand so the agent stays aligned with your design decisions.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;Archived&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;Right side panel (bottom)&lt;/td&gt;
&lt;td&gt;Completed or superseded documents from earlier steps (e.g., &lt;em&gt;Phase 1 Plan&lt;/em&gt; once Phase 1 is finished).&lt;/td&gt;
&lt;td&gt;Kept accessible for historical lookup, but separated from active work so agents do not waste context tokens on finished phases.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;blockquote&gt;
&lt;p&gt;&lt;strong&gt;Tip: Let your AI tidy the session for you&lt;/strong&gt;&lt;br&gt;
When &lt;code&gt;Working&lt;/code&gt; gets crowded, you do not have to move every card manually. Just tell your connected agent in chat:&lt;br&gt;
&lt;em&gt;"Move the API Spec artifact to References and the completed Phase 1 Plan to this session's Archived zone."&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Good to know:&lt;/strong&gt; Moving a card into a Session's &lt;strong&gt;Archived&lt;/strong&gt; zone only organizes that session's layout—it does &lt;strong&gt;not&lt;/strong&gt; globally archive the artifact or schedule it for deletion. Active (unarchived) artifacts and sessions are kept indefinitely in &lt;code&gt;~/.ai-artifacts/&lt;/code&gt;; only items explicitly moved to the global &lt;strong&gt;Archive&lt;/strong&gt; are automatically cleaned up after &lt;strong&gt;90 days&lt;/strong&gt; (&lt;code&gt;agentPlus.artifactRetentionDays&lt;/code&gt;).&lt;/p&gt;
&lt;/blockquote&gt;

&lt;h3&gt;
  
  
  Write Session Instructions (Your Agent's Onboarding Brief)
&lt;/h3&gt;

&lt;p&gt;In the right-hand panel under &lt;strong&gt;Session Type&lt;/strong&gt;, you will find the &lt;strong&gt;Instructions&lt;/strong&gt; block:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Type your project rules, coding conventions, or which &lt;code&gt;References&lt;/code&gt; documents an agent should read before starting work, then click &lt;strong&gt;Save&lt;/strong&gt; (or ask your AI in chat to help draft and save the Instructions with you).&lt;/li&gt;
&lt;li&gt;Prefer a full editor tab? Click &lt;strong&gt;Open Markdown File&lt;/strong&gt; to edit the session's &lt;code&gt;instructions.md&lt;/code&gt; file directly in VS Code—changes sync back to the Session Manager automatically.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;When an agent joins the session, it keeps these &lt;strong&gt;Instructions&lt;/strong&gt; in memory and checks them before starting related tasks.&lt;/p&gt;

&lt;h3&gt;
  
  
  Connect a fresh chat or another AI client
&lt;/h3&gt;

&lt;p&gt;When your current chat thread gets long—or when you want to open a second chat to continue the next phase—you do not need to re-explain your setup:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Click &lt;strong&gt;&lt;code&gt;Connect&lt;/code&gt;&lt;/strong&gt; in the &lt;strong&gt;Work Session Manager&lt;/strong&gt; header (or from the session's item in the &lt;strong&gt;Sessions&lt;/strong&gt; sidebar) to copy the join prompt.&lt;/li&gt;
&lt;li&gt;Open a &lt;strong&gt;new chat&lt;/strong&gt; in any supported AI client and paste the prompt with your next task:
&lt;/li&gt;
&lt;/ol&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;[Pasted Session Connect Prompt]
Read the Architecture Blueprint in References, then create an implementation-plan artifact for Phase 2 so I can review it.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;The new agent joins the session, reads your &lt;strong&gt;Instructions&lt;/strong&gt; and the categorized document catalog, consults the blueprint in &lt;code&gt;References&lt;/code&gt;, creates the new plan in &lt;strong&gt;&lt;code&gt;Working&lt;/code&gt;&lt;/strong&gt;, and opens the &lt;strong&gt;Preview UI&lt;/strong&gt; for your review.&lt;/p&gt;




&lt;h2&gt;
  
  
  Handy Session Controls
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Pause Auto-Open Temporarily:&lt;/strong&gt; In the Session Manager header, open the &lt;strong&gt;Settings (gear)&lt;/strong&gt; menu and choose &lt;strong&gt;Turn off auto-open artifact&lt;/strong&gt; if you want agents to create or update artifacts in the background without interrupting your reading.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Export a Portable ZIP:&lt;/strong&gt; From the &lt;strong&gt;Settings (gear)&lt;/strong&gt; menu, choose &lt;strong&gt;Export&lt;/strong&gt; to save a &lt;code&gt;.zip&lt;/code&gt; archive containing &lt;code&gt;Instructions.md&lt;/code&gt; and organized &lt;code&gt;Working/&lt;/code&gt;, &lt;code&gt;References/&lt;/code&gt;, and &lt;code&gt;Archived/&lt;/code&gt; folders with numbered Markdown files—ready to share with teammates or store outside the IDE.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Find Sessions Quickly from the Sidebar:&lt;/strong&gt; In the &lt;strong&gt;AI Artifacts&lt;/strong&gt; view in the Activity Bar under &lt;strong&gt;Sessions&lt;/strong&gt;, you can &lt;strong&gt;Pin&lt;/strong&gt; active sessions to the top, filter by &lt;strong&gt;Session Type&lt;/strong&gt; (&lt;code&gt;Conversation&lt;/code&gt;, &lt;code&gt;Project&lt;/code&gt;, &lt;code&gt;Feature&lt;/code&gt;, &lt;code&gt;Fix&lt;/code&gt;, or custom types), or search sessions by title.&lt;/li&gt;
&lt;/ul&gt;




&lt;h2&gt;
  
  
  What's next?
&lt;/h2&gt;

&lt;p&gt;Once your project context and instructions live in a &lt;strong&gt;Work Session&lt;/strong&gt;, you can hand work off across different AI coding clients—or let multiple agents coordinate automatically:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Workflow overview:&lt;/strong&gt; &lt;a href="https://dev.to/huynguyen294/tutorial-00-ai-artifacts-workflow-guide-when-to-use-what-2hjf"&gt;Part 00: AI Artifacts Workflow Guide — When to Use What&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Next guide:&lt;/strong&gt; &lt;a href="https://dev.to/huynguyen294/cross-client-handoffs-reusable-context-across-ai-agents-40d9"&gt;Part 03: Cross-Client Handoffs &amp;amp; Reusable Context Across AI Agents&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Automated workflows:&lt;/strong&gt; &lt;a href="https://dev.to/huynguyen294/automated-multi-agent-workflows-43mb"&gt;Part 04: Automated Multi-Agent Workflows with Session Waiters &amp;amp; Waiting Room&lt;/a&gt;
&lt;/li&gt;
&lt;/ul&gt;

</description>
      <category>vscode</category>
      <category>ai</category>
      <category>agents</category>
      <category>mcp</category>
    </item>
    <item>
      <title>Built an MCP extension that lets AI coding agents create visual artifacts directly in your IDE. Would love to hear your feedback!</title>
      <dc:creator>Huy Nguyen</dc:creator>
      <pubDate>Wed, 07 Oct 2026 11:49:49 +0000</pubDate>
      <link>https://dev.to/huynguyen294/built-an-mcp-extension-that-lets-ai-coding-agents-create-visual-artifacts-directly-in-your-ide-4jdf</link>
      <guid>https://dev.to/huynguyen294/built-an-mcp-extension-that-lets-ai-coding-agents-create-visual-artifacts-directly-in-your-ide-4jdf</guid>
      <description>&lt;div class="ltag__link--embedded"&gt;
  &lt;div class="crayons-story "&gt;
  &lt;a href="https://dev.to/huynguyen294/this-extension-lets-your-ai-agent-create-artifacts-for-plans-workarounds-and-designs-in-cursor-19n3" class="crayons-story__hidden-navigation-link"&gt;This extension lets your AI agent create artifacts for plans, workarounds and designs in Cursor, Codex and GitHub Copilot&lt;/a&gt;


  &lt;div class="crayons-story__body crayons-story__body-full_post"&gt;
    &lt;div class="crayons-story__top"&gt;
      &lt;div class="crayons-story__meta"&gt;
        &lt;div class="crayons-story__author-pic"&gt;

          &lt;a href="/huynguyen294" class="crayons-avatar  crayons-avatar--l  "&gt;
            &lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F1040842%2F7ff7476a-b560-4cd4-b82e-c72f90603057.jpeg" alt="huynguyen294 profile" class="crayons-avatar__image"&gt;
          &lt;/a&gt;
        &lt;/div&gt;
        &lt;div&gt;
          &lt;div&gt;
            &lt;a href="/huynguyen294" class="crayons-story__secondary fw-medium m:hidden"&gt;
              Huy Nguyen
            &lt;/a&gt;
            &lt;div class="profile-preview-card relative mb-4 s:mb-0 fw-medium hidden m:inline-block"&gt;
              
                Huy Nguyen
                
                
              
              &lt;div id="story-author-preview-content-4811067" class="profile-preview-card__content crayons-dropdown branded-7 p-4 pt-0"&gt;
                &lt;div class="gap-4 grid"&gt;
                  &lt;div class="-mt-4"&gt;
                    &lt;a href="/huynguyen294" class="flex"&gt;
                      &lt;span class="crayons-avatar crayons-avatar--xl mr-2 shrink-0"&gt;
                        &lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F1040842%2F7ff7476a-b560-4cd4-b82e-c72f90603057.jpeg" class="crayons-avatar__image" alt=""&gt;
                      &lt;/span&gt;
                      &lt;span class="crayons-link crayons-subtitle-2 mt-5"&gt;Huy Nguyen&lt;/span&gt;
                    &lt;/a&gt;
                  &lt;/div&gt;
                  &lt;div class="print-hidden"&gt;
                    
                      Follow
                    
                  &lt;/div&gt;
                  &lt;div class="author-preview-metadata-container"&gt;&lt;/div&gt;
                &lt;/div&gt;
              &lt;/div&gt;
            &lt;/div&gt;

          &lt;/div&gt;
          &lt;a href="https://dev.to/huynguyen294/this-extension-lets-your-ai-agent-create-artifacts-for-plans-workarounds-and-designs-in-cursor-19n3" class="crayons-story__tertiary fs-xs"&gt;&lt;time&gt;Oct 7&lt;/time&gt;&lt;span class="time-ago-indicator-initial-placeholder"&gt;&lt;/span&gt;&lt;/a&gt;
        &lt;/div&gt;
      &lt;/div&gt;

    &lt;/div&gt;

    &lt;div class="crayons-story__indention"&gt;
      &lt;h2 class="crayons-story__title crayons-story__title-full_post"&gt;
        &lt;a href="https://dev.to/huynguyen294/this-extension-lets-your-ai-agent-create-artifacts-for-plans-workarounds-and-designs-in-cursor-19n3" id="article-link-4811067"&gt;
          This extension lets your AI agent create artifacts for plans, workarounds and designs in Cursor, Codex and GitHub Copilot
        &lt;/a&gt;
      &lt;/h2&gt;
        &lt;div class="crayons-story__tags"&gt;
            &lt;a class="crayons-tag  crayons-tag--monochrome " href="/t/vscode"&gt;&lt;span class="crayons-tag__prefix"&gt;#&lt;/span&gt;vscode&lt;/a&gt;
            &lt;a class="crayons-tag  crayons-tag--monochrome " href="/t/ai"&gt;&lt;span class="crayons-tag__prefix"&gt;#&lt;/span&gt;ai&lt;/a&gt;
            &lt;a class="crayons-tag  crayons-tag--monochrome " href="/t/mcp"&gt;&lt;span class="crayons-tag__prefix"&gt;#&lt;/span&gt;mcp&lt;/a&gt;
            &lt;a class="crayons-tag  crayons-tag--monochrome " href="/t/cursor"&gt;&lt;span class="crayons-tag__prefix"&gt;#&lt;/span&gt;cursor&lt;/a&gt;
        &lt;/div&gt;
      &lt;div class="crayons-story__bottom"&gt;
        &lt;div class="crayons-story__details"&gt;
          &lt;a href="https://dev.to/huynguyen294/this-extension-lets-your-ai-agent-create-artifacts-for-plans-workarounds-and-designs-in-cursor-19n3" class="crayons-btn crayons-btn--s crayons-btn--ghost crayons-btn--icon-left"&gt;
            &lt;div class="multiple_reactions_aggregate"&gt;
              &lt;span class="multiple_reactions_icons_container"&gt;
                  &lt;span class="crayons_icon_container"&gt;
                    &lt;img src="https://assets.dev.to/assets/sparkle-heart-5f9bee3767e18deb1bb725290cb151c25234768a0e9a2bd39370c382d02920cf.svg" width="18" height="18"&gt;
                  &lt;/span&gt;
              &lt;/span&gt;
              &lt;span class="aggregate_reactions_counter"&gt;2&lt;span class="hidden s:inline"&gt;&amp;nbsp;reactions&lt;/span&gt;&lt;/span&gt;
            &lt;/div&gt;
          &lt;/a&gt;
            &lt;a href="https://dev.to/huynguyen294/this-extension-lets-your-ai-agent-create-artifacts-for-plans-workarounds-and-designs-in-cursor-19n3#comments" class="crayons-btn crayons-btn--s crayons-btn--ghost crayons-btn--icon-left flex items-center"&gt;
              

              &lt;span class="hidden s:inline"&gt;Add&amp;nbsp;Comment&lt;/span&gt;
            &lt;/a&gt;
        &lt;/div&gt;
        &lt;div class="crayons-story__save"&gt;
          &lt;small class="crayons-story__tertiary fs-xs mr-2"&gt;
            6 min read
          &lt;/small&gt;
          
          &lt;div class="only-sidebar-menu-item trusted-visible-block print-hidden align-center"&gt;
                
                    
                
          &lt;/div&gt;

        &lt;/div&gt;
      &lt;/div&gt;
    &lt;/div&gt;
  &lt;/div&gt;
&lt;/div&gt;


&lt;/div&gt;


</description>
      <category>ai</category>
      <category>mcp</category>
      <category>programming</category>
    </item>
    <item>
      <title>This extension lets your AI agent create artifacts for plans, workarounds and designs in Cursor, Codex and GitHub Copilot</title>
      <dc:creator>Huy Nguyen</dc:creator>
      <pubDate>Wed, 07 Oct 2026 11:45:35 +0000</pubDate>
      <link>https://dev.to/huynguyen294/this-extension-lets-your-ai-agent-create-artifacts-for-plans-workarounds-and-designs-in-cursor-19n3</link>
      <guid>https://dev.to/huynguyen294/this-extension-lets-your-ai-agent-create-artifacts-for-plans-workarounds-and-designs-in-cursor-19n3</guid>
      <description>&lt;p&gt;This guide uses an extension called AI Artifacts to let &lt;strong&gt;Cursor, Claude Code, Codex, Windsurf, GitHub Copilot, Cline, or Antigravity&lt;/strong&gt; produce a plan, workaround note, or design as a separate document—an artifact—inside your IDE. Instead of scrolling through a narrow chat sidebar or reading raw Markdown syntax, you read a rendered preview (with formatted tables, syntax-highlighted code, and Mermaid diagrams), comment on exact passages, and send all feedback back in a single step.&lt;/p&gt;

&lt;h2&gt;
  
  
  What you will need
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;VS Code 1.95 or newer, or a compatible editor such as Cursor, Windsurf, Antigravity or VSCodium&lt;/li&gt;
&lt;li&gt;An AI agent that supports MCP and skills (Cursor, Claude Code, Codex, Windsurf, GitHub Copilot, Cline or Antigravity)&lt;/li&gt;
&lt;li&gt;Node.js available as &lt;code&gt;node&lt;/code&gt; in your &lt;code&gt;PATH&lt;/code&gt;
&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Step 1: Install and connect the extension
&lt;/h2&gt;

&lt;p&gt;Install &lt;strong&gt;AI Artifacts&lt;/strong&gt; from the &lt;a href="https://marketplace.visualstudio.com/items?itemName=huynguyen294.ai-artifacts" rel="noopener noreferrer"&gt;VS Code Marketplace&lt;/a&gt; or &lt;a href="https://open-vsx.org/extension/huynguyen294/ai-artifacts" rel="noopener noreferrer"&gt;Open VSX&lt;/a&gt;. The extension communicates with your agent through the Model Context Protocol (MCP) and a small skill that defines how documents are created and updated.&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;On first activation, click &lt;strong&gt;Run Installer&lt;/strong&gt; in the prompt—or open the Command Palette (&lt;code&gt;Ctrl+Shift+P&lt;/code&gt; / &lt;code&gt;Cmd+Shift+P&lt;/code&gt;) and run &lt;strong&gt;AI Artifacts: Installer&lt;/strong&gt;. This preselects detected clients on the machine and leaves unrelated MCP settings and skills unchanged; press &lt;code&gt;Enter&lt;/code&gt; (or click &lt;strong&gt;OK&lt;/strong&gt;) to configure them.&lt;/li&gt;
&lt;li&gt;Click &lt;strong&gt;Reload Window&lt;/strong&gt; when prompted.&lt;/li&gt;
&lt;li&gt;Restart your AI chat extension or agent, then start a &lt;strong&gt;new&lt;/strong&gt; chat. A chat that was already open cannot load tools installed after it started.&lt;/li&gt;
&lt;li&gt;Run &lt;strong&gt;AI Artifacts: Verify All Integrations&lt;/strong&gt;. The skill and the base runtime should both report &lt;strong&gt;Ready&lt;/strong&gt;.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fyxpvhyccb075oax2k6qi.gif" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fyxpvhyccb075oax2k6qi.gif" alt="_Running AI Artifacts: Installer_" width="720" height="414"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  Step 2: Ask the agent to create an artifact
&lt;/h2&gt;

&lt;p&gt;The agent creates a review document only when it is asked to. A request for "a plan" on its own produces a plan in the chat. Include the word &lt;em&gt;artifact&lt;/em&gt; in the prompt:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;Create a dummy plan with: a mermaid diagram, a comparison table, code blocks,
a task list, and open questions. Create an artifact for me to review.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;You can also name the skill directly: &lt;code&gt;create-review-artifact&lt;/code&gt; (in Codex, &lt;code&gt;$create-review-artifact&lt;/code&gt;).&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fez8rjtqz40inj5pkkc33.gif" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fez8rjtqz40inj5pkkc33.gif" alt="_The agent creates the artifact and the review editor opens_" width="759" height="427"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;The review editor opens in your focused IDE window automatically, and the agent also shares a clickable link to the artifact in chat so you can open it directly.&lt;/p&gt;

&lt;h2&gt;
  
  
  Step 3: Review the document with inline comments
&lt;/h2&gt;

&lt;p&gt;Read the plan in the rendered preview—with formatted tables, code blocks, and Mermaid diagrams instead of raw Markdown symbols. Select any paragraph, heading, list item, quote, code block or table cell, type a comment in the popover, and press &lt;code&gt;Enter&lt;/code&gt; to save it (&lt;code&gt;Shift+Enter&lt;/code&gt; inserts a new line). Saved comments are listed in the &lt;strong&gt;Comments&lt;/strong&gt; drawer, and the &lt;strong&gt;Review&lt;/strong&gt; button shows how many are pending.&lt;/p&gt;

&lt;p&gt;In the example plan, four comments were left:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Interface sketch:&lt;/strong&gt; add more detail.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Tasks:&lt;/strong&gt; add more detail.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Acceptance criteria:&lt;/strong&gt; delete the section, which is unnecessary for a feature this small.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Rollback:&lt;/strong&gt; delete the section for the same reason.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fq5ckyrp5t3xvv48oz5c8.gif" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fq5ckyrp5t3xvv48oz5c8.gif" alt="_Selecting text and leaving a comment_" width="759" height="427"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Comments can also be questions, such as asking why a particular approach was chosen.&lt;/p&gt;

&lt;h2&gt;
  
  
  Step 4: Submit the review
&lt;/h2&gt;

&lt;p&gt;Four decisions are available in the editor header:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Review&lt;/strong&gt; sends all saved comments to the agent and starts the next round. It requires at least one comment.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Proceed&lt;/strong&gt; approves the document. For a plan, the agent begins carrying it out immediately.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Just save&lt;/strong&gt; ends the review without executing anything. It suits cases where you only want to keep an idea: the agent saves the document to a file so you do not have to copy it manually.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Copy Markdown&lt;/strong&gt; copies the raw text to the clipboard without changing anything.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The example plan still needed changes, so &lt;strong&gt;Review&lt;/strong&gt; is the correct choice. The agent acknowledges the comments in the chat, summarizes what it is applying, and opens round 2.&lt;/p&gt;

&lt;h2&gt;
  
  
  Step 5: Verify round 2
&lt;/h2&gt;

&lt;p&gt;The agent answers questions in the chat and revises the document only where changes were requested. If a review contains only questions, the document stays byte-for-byte identical and the round still advances.&lt;/p&gt;

&lt;p&gt;Round 1 of the example, with four comments saved:&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fj64zv0vior24or6tftfe.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fj64zv0vior24or6tftfe.png" alt="_Round 1: four comments saved. Tasks, Acceptance criteria and Rollback are highlighted; the Interface sketch comment is just above_" width="800" height="747"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Round 2, after the agent applied them:&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fh01x5md6la03uqufuitf.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fh01x5md6la03uqufuitf.png" alt="_Round 2: a longer, more detailed task list, and no acceptance criteria or rollback sections_" width="800" height="747"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;The task list grew from 7 items to 11 with more concrete steps, the API example gained its response, and the two sections flagged for deletion were removed. The review can continue for as many rounds as needed, ending with &lt;strong&gt;Proceed&lt;/strong&gt; or &lt;strong&gt;Just save&lt;/strong&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  Step 6: Resume a review
&lt;/h2&gt;

&lt;p&gt;An artifact remains available after its editor tab is closed or the chat ends, so a review can be resumed without starting over.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;In the same chat&lt;/strong&gt;, ask the agent to reopen the artifact.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;In a new chat:&lt;/strong&gt;&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Run &lt;strong&gt;AI Artifacts: Search Artifact&lt;/strong&gt; from the Command Palette and type part of the title.&lt;/li&gt;
&lt;li&gt;Select the artifact to open it in the review editor.&lt;/li&gt;
&lt;li&gt;Click &lt;strong&gt;Inspect artifact&lt;/strong&gt; in the editor header. This copies a prompt that points to that exact artifact.&lt;/li&gt;
&lt;li&gt;Paste the prompt into the new chat.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;The agent reads the saved state and reports the current review round. Reconnecting never repeats an action that was already approved.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fyds8xtbermeec9k4was0.gif" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fyds8xtbermeec9k4was0.gif" alt="_Search an artifact, copy the Inspect prompt, paste it into a new chat, and the agent reads the saved state_" width="599" height="337"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;You can also browse and reopen recent documents directly from the &lt;strong&gt;AI Artifacts&lt;/strong&gt; sidebar in the Activity Bar under &lt;strong&gt;Recent Artifacts&lt;/strong&gt;.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fjis6xzotqldmlm16d49k.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fjis6xzotqldmlm16d49k.png" alt="_Reopening an artifact from Recent Artifacts in the AI Artifacts sidebar_" width="519" height="759"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  Step 7: Have a second AI review your plan
&lt;/h2&gt;

&lt;p&gt;Because &lt;strong&gt;Inspect artifact&lt;/strong&gt; lets any supported AI agent read the exact same artifact, you can get a second opinion on an important plan before you click &lt;strong&gt;Proceed&lt;/strong&gt;:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;While viewing the plan drafted by Agent 1 (for example, in Cursor), click &lt;strong&gt;Inspect artifact&lt;/strong&gt; in the editor header.&lt;/li&gt;
&lt;li&gt;Open a chat with a second AI agent (for example, Claude Code or Codex), paste the prompt, and ask it to review the proposal:
&lt;/li&gt;
&lt;/ol&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;   [Pasted Inspect artifact prompt]
   Review this plan for edge cases, missing steps, or simpler alternatives, and write your feedback into the artifact.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;ol&gt;
&lt;li&gt;Agent 2 reads the artifact and writes its review comments directly onto the document (or replies in chat), so you and Agent 1 can refine the plan in the next round before executing code.&lt;/li&gt;
&lt;/ol&gt;

&lt;h2&gt;
  
  
  Step 8: Use artifacts for other document types as well
&lt;/h2&gt;

&lt;p&gt;Artifacts are not limited to plans. You can use the same review workflow for requirements, designs, specs, analysis reports, or workaround notes:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;Create a requirements artifact for the webhook retry policy, including edge cases
and acceptance criteria, so I can review it.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;You can also ask the agent to create a &lt;code&gt;design&lt;/code&gt;, &lt;code&gt;spec&lt;/code&gt;, &lt;code&gt;analysis&lt;/code&gt;, &lt;code&gt;report&lt;/code&gt;, or &lt;code&gt;workaround&lt;/code&gt; artifact whenever you want to review structured documents outside the chat thread.&lt;/p&gt;

&lt;h2&gt;
  
  
  What's next?
&lt;/h2&gt;

&lt;p&gt;Reviewing a single artifact is all you need when a task revolves around one document. When a session produces multiple artifacts that you want to browse in one place—or when you want to carry shared instructions and context into fresh chats—move up to &lt;strong&gt;Work Sessions&lt;/strong&gt;:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Workflow overview:&lt;/strong&gt; &lt;a href="https://dev.to/huynguyen294/tutorial-00-ai-artifacts-workflow-guide-when-to-use-what-2hjf"&gt;Part 00: AI Artifacts Workflow Guide — When to Use What&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Next guide:&lt;/strong&gt; &lt;a href="https://dev.to/huynguyen294/organizing-work-sessions-from-simple-collections-to-shared-context-hlk"&gt;Part 02: Organizing Work Sessions — From Simple Collections to Agent Context&lt;/a&gt;
&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Links
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;&lt;a href="https://marketplace.visualstudio.com/items?itemName=huynguyen294.ai-artifacts" rel="noopener noreferrer"&gt;VS Code Marketplace&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href="https://open-vsx.org/extension/huynguyen294/ai-artifacts" rel="noopener noreferrer"&gt;Open VSX&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href="https://github.com/huynguyen294/ai-artifacts" rel="noopener noreferrer"&gt;GitHub&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Feedback is welcome in the comments or as an issue on GitHub. Please include the editor and agent you are using.&lt;/p&gt;

</description>
      <category>vscode</category>
      <category>ai</category>
      <category>mcp</category>
      <category>cursor</category>
    </item>
  </channel>
</rss>
