explainx / blog
Jul 16, 2026: Claude Code artifacts can call MCP connectors on every page load — Pro/Max/Team/Enterprise, viewer OAuth, no public links. explainx.ai maps setup, security, and Thariq thick-artifact patterns.
Jun 25, 2026
Released June 25, 2026, Claude Code 2.1.191 brings /rewind to undo /clear and restore prior context, fixes background agents restarting after stop, coalesces streaming updates for ~37% CPU savings, and patches comma-separated hook matchers that silently never fired.
Jul 13, 2026
@Saboo_Shubham_ reposted OpenAI's official Codex plugin for Claude Code — 28k+ GitHub stars, months old, not a surprise drop. explainx.ai maps /codex:review, /codex:rescue, transfer flows, and why Claude→Codex works but not the reverse.
Jul 12, 2026
@ClaudeDevs ships Claude Code desktop browser — read, click, debug URLs sandboxed. Version 1.2581.0 July 10. explainx.ai setup, shortcuts, and developer reactions.
On July 16, 2026, Anthropic's ClaudeDevs account posted that Claude Code artifacts can now call MCP connectors — a change that turned a ~495.9K-view thread into the week's most practical question for anyone building internal dashboards from coding sessions:
"Build an artifact once, then it pulls live data via MCP. Artifacts use the viewers' own MCP connectors, so creators don't need to re-run sessions to refresh, and viewers can only see the data they have access to."
That second line is the architectural shift. Before this ship, an artifact was a snapshot: whatever data Claude pulled while building the page was baked into HTML. Now the page is a thick artifact in Thariq Shihipar's sense — a durable shell that rehydrates from live systems every time someone opens it — without the publisher re-running Claude Code or leaking their credentials to viewers.
If you are new to the protocol layer, start with explainx.ai's MCP architecture guide. For the baseline Artifacts feature (sharing, org scopes, HTML constraints), see Claude Code Artifacts: shareable sessions.
| Question | Direct answer |
|---|---|
| What shipped? | Published artifacts can call remote MCP connectors on page load — live PR lists, issue boards, deploy metrics — not just build-time snapshots. |
| Which plans? | Pro, Max, Team, Enterprise. Not Free. Requires Claude Code v2.1.209+ (or desktop app 1.13576.0+). |
| Who authenticates? | The viewer, not the publisher. Each person uses their own claude.ai connector OAuth. |
| Public links? | No for MCP-backed artifacts — org-private or author-private only. |
Local .mcp.json servers? | Work during build only. Published pages call claude.ai remote connectors only. |
| Admin kill switch? | Enable artifact connectors toggle under Settings > Capabilities (Team/Enterprise Owners). |
| How does this relate to Thariq's framework? | Live MCP turns artifacts from static exports into thick artifacts + context — durable UI with fresh backend data. |
Claude Code Artifacts already let you publish a single HTML page from a session — annotated PR diffs, comparison grids, investigation timelines. The limitation was freshness: if your GitHub open-PR count changed after publish, the artifact did not know unless you asked Claude to republish.
Anthropic's official docs now describe connector-backed pages under Share session output as artifacts:
The ClaudeDevs announcement framed the builder workflow: "letting you build dashboards and apps that can fetch information and take actions for each viewer on demand." Actions matter — a page can expose controls that invoke connector tools with side effects (post a Slack message, update a Linear issue), and those actions run through whoever clicked the control, not the original author.
This is the piece most social threads gloss over. The artifact is a static HTML page under a strict Content Security Policy — no fetch, no WebSockets, no external scripts. The only outbound data path when someone views the page is handing connector calls to claude.ai, which makes the network request on the page's behalf.
Anthropic documents four rules:
| Rule | What it means in practice |
|---|---|
| Viewer credentials | Opening a GitHub-backed PR dashboard uses your GitHub connector token, not the author's. |
| Permission on first call | claude.ai prompts each viewer before the artifact's first connector invocation for that session. Preferences persist for subsequent uses of that artifact. |
| No credential exposure | The HTML never receives OAuth tokens — only structured tool results. |
| Side effects follow the clicker | A "Close issue" button in the artifact closes the issue as the viewer who clicked it. |
That model is why public sharing is disabled for connector-backed artifacts. A public URL has no authenticated viewer to broker OAuth through. Team and Enterprise org sharing works because every viewer signs into claude.ai as an org member.
For enterprise connector setup — who adds org-wide connectors vs who individually authenticates — see How to use Claude connectors and MCP servers. Note that org-level connector credential sharing on claude.ai remains a separate debate from artifact viewer scoping; artifact calls are explicitly per-viewing-account even when the connector was added at org level.
Hours before the ClaudeDevs MCP post, Thariq Shihipar (@trq212) posted his thin prompts, thick artifacts, thin skills framework — ~82K views on the same day. explainx.ai covered it in Thin prompts, thick artifacts, thin skills.
The connection is direct:
| Layer | Without MCP connectors | With MCP connectors |
|---|---|---|
| Thin prompt | "Build a PR dashboard artifact." | Same — intent only. |
| Thick artifact | Static HTML with PR data from the build session. | HTML shell + live PR list on every open. |
| Context | Codebase + session history while building. | Plus viewer's GitHub permissions at read time. |
Thariq has advocated HTML over Markdown for agent output since his unreasonable effectiveness of HTML guide — tables, CSS, interactive controls, shareability. MCP-backed artifacts are the next step: the HTML file stays the durable spec; connectors supply the moving parts.
Thin prompt example (pair with a thick artifact, not a paragraph of requirements):
Build a dashboard artifact of our open pull requests. Pull the live list through my GitHub connector when the page loads. Add a fallback message if the viewer hasn't connected GitHub.
Operational dashboard variant:
Build an artifact of last week's deploy failures by service. Refresh from my monitoring connector every 5 minutes. Include a manual refresh button.
Action-capable board (viewer-scoped writes):
Make a triage board artifact with open Linear issues as cards. Pull live issues on load. Let viewers move a card to Done — that should call the Linear connector to update status using their account.
Claude Code sessions can use two different MCP surfaces, and conflating them breaks deployments.
| Surface | Where configured | Available when | Use case |
|---|---|---|---|
Local MCP (.mcp.json, claude mcp add) | Your machine | Session only | Query Postgres while authoring the page; screenpipe timeline during build |
| claude.ai remote connectors | Settings > Connectors | Build and published artifact views | GitHub, Linear, Slack, custom remote MCP servers |
Local servers like Desktop Commander or context-mode remain powerful for authoring — Claude can pull real data into the first draft. But the published URL cannot call them. Plan connector allowlists around remote connectors your viewers will also have.
For wiring local servers into Claude Code sessions, see Claude Code MCP servers: connect any tool.
Connector calls are an exception to the artifact CSP — everything else is unchanged. From Anthropic's Page constraints:
Token cost also rises: styled interactive pages cost more output tokens than terminal text. Prefer SVG and CSS over inlined PNGs when building connector dashboards.
When the shell renders but live blocks stay blank, Anthropic lists three causes:
GitHub issues in the Claude Code tracker document additional edge cases from early Cowork artifact builds — for example, read-only SQL queries still triggering per-call Allow/Deny gates when execute_sql is classified as a write-capable tool. Treat connector tool permissions as per-artifact-context, not guaranteed to inherit global "Always Allow" settings.
Viewer-scoped auth solves credential leakage — viewers never inherit the publisher's tokens. It does not solve authorization design in your MCP server. If your custom connector returns all rows to any authenticated user, every viewer still sees all rows.
explainx.ai's MCP security guide still applies:
innerHTML without sanitization is a XSS surface; treat tool output as data.claude_artifact_* publish and share events; pair with your connector server's own invocation logs.Anthropic's connector authentication docs cover OAuth DCR, CIMD, and static header patterns for custom remote servers you want teammates to connect individually.
| Requirement | Detail |
|---|---|
| Plans | Pro, Max, Team, Enterprise |
| Claude Code CLI | v2.1.209 or later for connector calls at view time |
| Claude desktop app | v1.13576.0 or later |
| Authentication | Session signed in via /login — API-key-only sessions cannot publish artifacts |
| Public MCP artifacts | Not supported on any plan |
| Pro/Max sharing | Connector artifacts stay private to you (public links are otherwise the only Pro/Max share path) |
| Team/Enterprise sharing | Org members with the link, after connecting required connectors |
Earlier Claude Code versions still publish artifacts, but pages show build-time data only — no live connector refresh.
Artifacts remain session captures, not hosted applications. They cannot replace:
For full-stack generation with external hosting, dedicated builders still win. Artifacts win when the audience is your org, the spec lived in a Claude Code session, and each viewer should see only what their connectors allow.
/login to claude.ai.Browse additional servers on explainx.ai's MCP server directory when evaluating what to connect before you publish.
Official docs: Claude Code artifacts · Artifact MCP section · Claude Help Center — artifacts and MCP · Connectors overview
Plan availability, version floors, and connector behavior are accurate as of July 16, 2026. Confirm current requirements on code.claude.com before shipping production dashboards.