Skip to content

armgit / mso

Public Imported from GitHub
0

Browser-based graphical shell and control plane for a Linux server you own — real terminal, file manager, live metrics and BYOK AI in one mobile-first workspace. Next.js 16 · React 19 · node-pty · Tailwind v4 · shadcn/ui. Public alpha, self-hosted.

refactor(mcp): dedupe ChatGPT CSP metadata

Rahman790507976b41
diff --git a/docs/ARCHITECTURE.md b/docs/ARCHITECTURE.md
index 3a5bfa82..c97d317a 100644
--- a/docs/ARCHITECTURE.md
+++ b/docs/ARCHITECTURE.md
@@ -118,7 +118,7 @@ through the app catch-all.
### 4.1 ChatGPT as an MSO presentation target
The web catch-all remains the authenticated cockpit route. ChatGPT does **not** frame that cockpit.
-Instead `render_mso_surface` binds the versioned MCP App resource `ui://mso/surface-v4.html`, which
+Instead `render_mso_surface` binds the versioned MCP App resource `ui://mso/surface-v5.html`, which
implements a smaller presentation router for `/`, `/monitor`, `/project`, `/diff`, `/browser` and
`/apps/<reviewed-id>`. Native Surface views call the same bounded MSO tools; reviewed app demos may
use a nested iframe only when their exact HTTPS origin exists in the code-owned Surface catalog and
diff --git a/docs/CHANGELOG.md b/docs/CHANGELOG.md
index e2d7da80..8888c776 100644
--- a/docs/CHANGELOG.md
+++ b/docs/CHANGELOG.md
@@ -4,6 +4,12 @@
Newest first. `docs/PROGRESS.md` is the source of truth for *why* a change was made;
this is the *what*, and it is what Settings → About shows as “What's new”.
+## 2026-09-04
+
+**Changed**
+
+- `mcp` dedupe ChatGPT CSP metadata
+
## 2026-09-03
**Added**
diff --git a/docs/PROGRESS.md b/docs/PROGRESS.md
index 288cd909..3af3c00a 100644
--- a/docs/PROGRESS.md
+++ b/docs/PROGRESS.md
@@ -1,4 +1,5 @@
# mso — Progress Log
+- **MCP Apps CSP metadata dedup / Surface v5 (2026-09-04):** MCP server `1.9.4` / toolset `2026.09.04.5` moves all standard widget CSP fields to the preferred `_meta.ui.csp` surface and leaves legacy `openai/widgetCSP` with only `redirect_domains` for `Open in MSO`. Workflow/Surface resources are cache-busted to `workflow-progress-v3` and `surface-v5`; the reviewed Play Together iframe remains exact-origin under `/embed`, so one external-iframe review advisory is expected from `ui.csp.frameDomains`. Scanner-facing Surface metadata explicitly names `/apps/play-together` and has a regression guard against cross-scope app examples.
- **MCP Apps presentation unification / Surface v4 (2026-09-04):** MCP server `1.9.3` / toolset `2026.09.04.4` retires the specialized `project-status-v2`, `project-diff-v2`, and `vps-status-v2` resources. `project_get`, `project_diff`, and `vps_status` are pure data tools again; `ui://mso/surface-v4.html` is the only general-purpose UI and uses the portable MCP Apps `tools/call` bridge for native monitor/project/diff views. `workflow-progress-v2` remains intentionally separate for live workflow polling. Play Together keeps the exact `game.rahmanef.com/embed` frame boundary, so the external-iframe review advisory remains expected only for the universal Surface.
- **Cognitive Runtime claim audit / public evidence boundary (2026-09-04):** revalidated source/main/origin and loopback/public production on the latest concurrent baseline, then reran the deterministic Cognitive Runtime/memory gates: routing recall 100%, active-schema reduction 96.6%, typed memory 8/8, P9 6/6 at 200 corrections, P10 lifecycle 8/8, lexical/local-semantic 4/6 vs 3/6, and bounded graph fixture 3/3. Raw P6/P7/P8 artifacts were independently re-hashed and their aggregate arithmetic recomputed; the canonical P7 two-run artifact is `mso-p7-repeat-full-2x9.json` (`a0c12961...`), not the similarly named older `mso-p7-repeat-full-2runs.json`. Canonical P7+P8 yields 5/5 perfect full runs for both MSO and Hermes and the documented 74.7% token / 12.8% average-latency / 34.1% p50 observed deltas on that exact nine-scenario matched corpus. P4/P5 exact tables are downgraded to historical release-reported metrics because this audit could not independently recover raw artifacts matching those exact values. README now removes the stale 29-tool ChatGPT count, removes the unsupported "most complete" positioning superlative, and states explicit no-universal-superiority/no-cost/no-cache/no-OpenClaw boundaries. Tool Forge executable evaluation is documented as fail-closed until its explicit local Docker sandbox image is provisioned; that image was absent during this audit.
- **Play Together secure iframe / Surface v3 (2026-09-04):** MCP server `1.9.2` / toolset `2026.09.04.3` upgrades `ui://mso/surface-v3.html` so Play Together can render live inside ChatGPT without relaxing the whole app. The Surface allowlists only `https://game.rahmanef.com`, starts it at `/embed`, revalidates the exact origin plus `/embed` path prefix in-browser, and retains a sandbox without popups/top-navigation. The Play Together deployment reciprocally keeps normal routes anti-frame and gives only `/embed` + `/embed/*` the exact `mso-ui.rahmanef.com` ancestor permission. Auth remains normal; no demo bypass or raw HTML/arbitrary URL input is added.
diff --git a/docs/TESTING-HANDOFF.md b/docs/TESTING-HANDOFF.md
index a5d5ea52..bfaf970a 100644
--- a/docs/TESTING-HANDOFF.md
+++ b/docs/TESTING-HANDOFF.md
@@ -5,12 +5,12 @@ Update this file at the end of every manual or automated test session. Keep it f
## Current status
-The current release target is **MSO CLI 1.12.0**, MCP server `1.9.2`, toolset `2026.09.04.3`; use `mso --version` plus `/api/health` for the exact Git build because documentation-only release commits also advance that identity. The v1.10/v1.11 findings and the v1.12 sectioned-terminal/recoverable-error work below are resolved and retained as historical test evidence. Do not re-open a resolved finding from prose alone: reproduce it against the current build first. There is no known 1.12 release blocker in this handoff.
+The current release target is **MSO CLI 1.12.0**, MCP server `1.9.4`, toolset `2026.09.04.5`; use `mso --version` plus `/api/health` for the exact Git build because documentation-only release commits also advance that identity. The v1.10/v1.11 findings and the v1.12 sectioned-terminal/recoverable-error work below are resolved and retained as historical test evidence. Do not re-open a resolved finding from prose alone: reproduce it against the current build first. There is no known 1.12 release blocker in this handoff.
Current verified behavior:
- **2026-09-04 Cognitive Runtime public-claim audit:** source/main/origin + loopback/public production were revalidated on the latest concurrent baseline. Fresh reruns pass `bench:cognitive`, `bench:memory`, P9 200-correction calibration, P10 lifecycle, and P10 retrieval calibration. Current deterministic evidence: 100% required-tool recall, 96.6% active-schema reduction, memory 8/8, P9 6/6, P10 lifecycle 8/8, lexical/local-semantic 4/6 vs 3/6, graph fixture 3/3. Canonical raw comparison artifacts were re-hashed: P6 `e0b5e1e30a9f1e5b34624e42d4bb5195792e50129b74de7d4006906a99716ba5`, P7 two-run `a0c12961d142420f6ac334ff021bde7ab82faa89f9294107d3bd46823606a530`, P8 three-run `f70e2f5e8132f0808976ed69e2cb6769d2cec861d58612a8cca2ef5d80a7d838`, P8 cache `b4a2b1be5dc6edfb2ce8282eae64e3a4fc206cd043ecb825a1eae6a617c9f871`. The canonical P7 file is `mso-p7-repeat-full-2x9.json`; the similarly named `mso-p7-repeat-full-2runs.json` is a different, non-canonical run. Recomputed P7+P8 evidence is 5/5 perfect for both MSO and Hermes with 74.7% fewer normalized tokens/attempt, 12.8% lower mean average latency, and 34.1% lower mean per-run p50 for MSO on that exact matched nine-scenario corpus. P4/P5 exact numeric tables are historical-only because their exact raw artifacts were not independently revalidated in this audit. Cost/cache/OpenClaw remain unranked. Tool Forge source/non-Docker tests pass, but the local Forge Docker image was absent, so three Docker-only tests were correctly skipped and executable evaluation fails closed until explicitly provisioned.
-- Fresh 3 current MCP contract: server `1.9.3` / toolset `2026.09.04.4`; ChatGPT receives **64 transport tools = 63 model/operator actions (34 read / 17 write / 12 exec) + app-only `workflow_status`**, while the full transport catalog is 88 tools / 87 model tools. Every ChatGPT action advertises `outputSchema`, the dedicated MCP Apps widget origin is `https://mso-ui.rahmanef.com`, and five official digest-verified Skills are exposed through the MCP Skills/resource extension. Exactly two MCP App resources are published: dedicated `ui://mso/workflow-progress-v2.html` plus universal `ui://mso/surface-v4.html`. `project_get`, `project_diff`, and `vps_status` are pure data tools with no UI template; Surface native views refresh them through MCP Apps `tools/call`. The Surface accepts no raw HTML/arbitrary URL; its only frame domain is `https://game.rahmanef.com`, and every Play Together iframe URL must remain under `/embed`. The normal Play Together shell stays anti-frame, while the dedicated embed namespace allows only `mso-ui.rahmanef.com`. Other MSO server scopes are not copied into this catalog.
-- MCP App navigation acceptance: a ChatGPT-opened root URL carrying `redirectUrl=https://chatgpt.com/c/...` must 307 to `/assistant/mcp`, then authenticated browser E2E must show the Alfa `MCP Activity` view. The dedicated workflow card targets `/assistant/mcp`; VPS/project/diff are now native routes inside universal Surface v4 rather than independent widgets. Surface v4 is a cache-busting resource URI change, so ChatGPT must re-scan Fresh 3 after deployment.
+- Fresh 3 current MCP contract: server `1.9.4` / toolset `2026.09.04.5`; ChatGPT receives **64 transport tools = 63 model/operator actions (34 read / 17 write / 12 exec) + app-only `workflow_status`**, while the full transport catalog is 88 tools / 87 model tools. Every ChatGPT action advertises `outputSchema`, the dedicated MCP Apps widget origin is `https://mso-ui.rahmanef.com`, and five official digest-verified Skills are exposed through the MCP Skills/resource extension. Exactly two MCP App resources are published: dedicated `ui://mso/workflow-progress-v3.html` plus universal `ui://mso/surface-v5.html`. `project_get`, `project_diff`, and `vps_status` are pure data tools with no UI template; Surface native views refresh them through MCP Apps `tools/call`. The Surface accepts no raw HTML/arbitrary URL; its only frame domain is `https://game.rahmanef.com`, and every Play Together iframe URL must remain under `/embed`. The normal Play Together shell stays anti-frame, while the dedicated embed namespace allows only `mso-ui.rahmanef.com`. Other MSO server scopes are not copied into this catalog.
+- MCP App navigation acceptance: a ChatGPT-opened root URL carrying `redirectUrl=https://chatgpt.com/c/...` must 307 to `/assistant/mcp`, then authenticated browser E2E must show the Alfa `MCP Activity` view. The dedicated workflow card targets `/assistant/mcp`; VPS/project/diff are native routes inside universal Surface v5 rather than independent widgets. Surface v5 plus workflow-progress-v3 are cache-busting resource URI changes, so ChatGPT must re-scan Fresh 3 after deployment.
- P10 memory retention/lifecycle: isolated fixtures now archive only cold finished claims before the hard 2,000-record / 2 MiB live-ledger limits. Trigger/target are 1,400 -> <=1,000 live records (or ~1.4 MiB -> ~1 MiB); archive segments are content-addressed, owner-only and archive-before-ledger atomic. `bench:memory:lifecycle` is **8/8 deterministic** with a 1,400-correction seed ending at **1,000 live + 401 archived / two segments** and a 56-byte resolved projection. It verifies explicit archived-history retrieval, historical temporal resolution, aggregate-only no-content telemetry, `0700/0600` permissions, interrupted-commit live-wins semantics, backdated replacement over immutable archive, and frozen session A / fresh session B / resume-copy behavior. The P10 retrieval gate is **4/6 lexical** (synonym + Indonesian/English misses remain explicit); P10D also calibrates the existing local `mso-local-hybrid-v1` encoder on the same rows and gets **3/6** with zero network calls / zero persisted memory vectors, so production vector storage stays blocked by measured non-improvement rather than assumption. **3/3 two-hop relationship fixtures** keep graph storage blocked. The exact-history crowding bug (`Runtime-1` vs `Runtime-10...`) is fixed by exact phrase/token scoring; legacy `bench:memory` remains 100%.
- P9 memory calibration: isolated temporary-memory fixtures now cover repeated corrections, explicit-vs-observed authority, confidence/recency, future validity, forget, and history integrity. P9 fixed a real temporal privacy bug where `forget` could leave a future-scheduled claim that resurrected later; forget now closes all current/future-capable intervals while preserving already-ended history. Final 200-correction stress from source `d249da4`: **6/6 calibration scenarios**, existing memory benchmark **100%**, Cognitive Runtime memory gates **100%**; 200 ledger rows / 135,180 bytes resolve to 1 row / 58-byte compatibility projection (**99.96% smaller**) and history retrieval remains capped at 100. No calibrated case requires graph memory.
- P8 calibration: repeat aggregation now reports descriptive per-run spread without inferential claims; three new complete nine-scenario runs from source `51ae31f` were **27/27 task + 27/27 scenario-policy and 3/3 perfect runs for both agents**, with 54/54 row identity/provider/model/token-arithmetic checks passing. Combined with P7, both agents are **5/5 perfect**; five-run means are MSO vs Hermes 16,907.6 vs 66,791.1 normalized tokens/attempt (~74.7% lower for MSO), 18,985.9 vs 21,779.8 ms average latency, and 14,130.8 vs 21,457.2 ms per-run p50. New `bench:cache-calibration` final 4-round/12K-prefix run was 8/8 success; MSO cache-read was zero on 4/4 while Hermes naturally reported positive cache reads on 2/4 (9,728 + 19,968), but cache frequency is unranked because request envelopes differ. P8 also fixes false token non-comparability when independently valid normalized proof shapes vary across cache/no-cache rows. Cost remains non-comparable because attribution is asymmetric.
diff --git a/docs/generated/MCP-CATALOG.md b/docs/generated/MCP-CATALOG.md
index 60f2259c..3530d52b 100644
--- a/docs/generated/MCP-CATALOG.md
+++ b/docs/generated/MCP-CATALOG.md
@@ -4,13 +4,13 @@
## Full MSO catalog
-<!-- mcp-toolset: server=1.9.3 version=2026.09.04.4 tools=87 read=43 write=25 exec=19 -->
+<!-- mcp-toolset: server=1.9.4 version=2026.09.04.5 tools=87 read=43 write=25 exec=19 -->
| Fact | Current source value |
|---|---:|
-| MCP server | `1.9.3` |
-| Toolset | `2026.09.04.4` |
-| Toolset changed at | `2026-09-03T20:56:00Z` |
+| MCP server | `1.9.4` |
+| Toolset | `2026.09.04.5` |
+| Toolset changed at | `2026-09-04T01:36:00Z` |
| Transport tools | **88** |
| Model/operator tools | **87** |
| Read | **43** |
@@ -120,7 +120,7 @@
## ChatGPT static profile
-<!-- mcp-chatgpt-profile: server=1.9.3 version=2026.09.04.4 tools=63 read=34 write=17 exec=12 app-only=1 total=64 -->
+<!-- mcp-chatgpt-profile: server=1.9.4 version=2026.09.04.5 tools=63 read=34 write=17 exec=12 app-only=1 total=64 -->
The ChatGPT profile is a fail-closed static projection defined by `CHATGPT_TOOL_NAMES`. OAuth scope is still enforced independently; project-owned MCP tool names remain dynamic data behind the generic project bridge.
diff --git a/docs/integrations/CHATGPT-UI.md b/docs/integrations/CHATGPT-UI.md
index 7797692b..2341a3bc 100644
--- a/docs/integrations/CHATGPT-UI.md
+++ b/docs/integrations/CHATGPT-UI.md
@@ -20,8 +20,8 @@ Official references:
| Experience | Entry tool | Resource | Refresh behavior |
| --- | --- | --- | --- |
-| Workflow progress | `workflow_start` | `ui://mso/workflow-progress-v2.html` | app-only `workflow_status` |
-| Universal MSO Surface | `render_mso_surface` | `ui://mso/surface-v4.html` | `/monitor`, `/project`, and `/diff` call pure data tools via `tools/call`; Play Together uses its reviewed exact-origin `/embed` frame boundary |
+| Workflow progress | `workflow_start` | `ui://mso/workflow-progress-v3.html` | app-only `workflow_status` |
+| Universal MSO Surface | `render_mso_surface` | `ui://mso/surface-v5.html` | `/monitor`, `/project`, and `/diff` call pure data tools via `tools/call`; Play Together uses its reviewed exact-origin `/embed` frame boundary |
Source boundaries:
@@ -74,6 +74,8 @@ All data tools keep portable structured/text results for non-MCP-Apps clients.
## Dedicated UI origin
+Both templates keep standard CSP exclusively in `_meta.ui.csp`. The legacy `_meta["openai/widgetCSP"]` compatibility object contains only `redirect_domains` for `Open in MSO`; it deliberately does not duplicate `connect_domains`, `resource_domains`, or `frame_domains`. This keeps the scanner-facing policy single-source while preserving ChatGPT's trusted `openExternal` redirect allowlist.
+
Both templates declare `_meta.ui.domain = https://mso-ui.rahmanef.com` and the ChatGPT compatibility alias `_meta["openai/widgetDomain"]` with the same origin. The sibling hostname is intentional: the cockpit may scope its session cookie to `mso.rahmanef.com`, while `mso-ui.rahmanef.com` is outside that cookie domain. The dedicated origin therefore satisfies the plugin UI identity requirement without reusing the authenticated dashboard origin.
The reproducible Traefik route is stored in `ops/traefik/mso-ui.yml` and forwards only to the same local MSO service through the existing HTTPS/Let's Encrypt ingress. Widget CSP still allows no network/static-resource domains and only permits `Open in MSO` to redirect to `https://mso.rahmanef.com`.
diff --git a/lib/mcp/tools-surface.test.ts b/lib/mcp/tools-surface.test.ts
index bfb8b2b6..cc50d213 100644
--- a/lib/mcp/tools-surface.test.ts
+++ b/lib/mcp/tools-surface.test.ts
@@ -25,6 +25,13 @@ describe("MSO Surface MCP tools", () => {
expect(result.app).not.toHaveProperty("sandbox");
});
+ it("keeps scanner-facing Surface metadata scoped to Play Together", () => {
+ const render = tool("render_mso_surface");
+ const metadata = JSON.stringify({ description: render.description, chatgptDescription: render.chatgptDescription });
+ expect(metadata).toContain("play-together");
+ expect(metadata.toLowerCase()).not.toContain("antinrml");
+ });
+
it("keeps project identity separate from the route string", async () => {
const result = await tool("render_mso_surface").run({ route: "/project", project: "mso" }, { scope: "read" });
expect(result).toMatchObject({ route: "/project", kind: "project", project: "mso", openPath: "/files" });
diff --git a/lib/mcp/tools-surface.ts b/lib/mcp/tools-surface.ts
index 1fecbb93..d9cf17f6 100644
--- a/lib/mcp/tools-surface.ts
+++ b/lib/mcp/tools-surface.ts
@@ -46,7 +46,7 @@ export const SURFACE_TOOLS: McpTool[] = [
name: "render_mso_surface",
title: "Render MSO Surface",
description: "Render the universal MSO Surface MCP App. Use an MSO-style route such as /, /monitor, /project, /diff, /browser, or /apps/<reviewed-app-id>. For project/diff views pass project separately. This tool never accepts raw HTML or arbitrary external URLs; demo origins are resolved only from the server-owned allowlist.",
- chatgptDescription: "Render a secure interactive MSO Surface in ChatGPT. Route examples: /, /monitor, /project, /diff, /browser, /apps/play-together. No arbitrary HTML/URL input is accepted.",
+ chatgptDescription: "Render the secure interactive MSO Surface in ChatGPT. Native routes: /, /monitor, /project, /diff, /browser; reviewed live demo: /apps/play-together. Raw HTML and arbitrary URLs are rejected.",
scope: "read", annotations: READ_ONLY,
inputSchema: S({
route: { type: "string", minLength: 1, maxLength: 1024, description: "MSO Surface route. External URLs are rejected." },
diff --git a/lib/mcp/toolset.ts b/lib/mcp/toolset.ts
index 4f6d5136..11a574ef 100644
--- a/lib/mcp/toolset.ts
+++ b/lib/mcp/toolset.ts
@@ -3,9 +3,9 @@ import type { Scope } from "./scope";
import type { McpTool } from "./tool-kit";
import { toolDescriptor, type McpToolProfile } from "./tool-contract";
-export const MCP_SERVER_VERSION = "1.9.3";
-export const MCP_TOOLSET_VERSION = "2026.09.04.4";
-export const MCP_TOOLSET_CHANGED_AT = "2026-09-03T20:56:00Z";
+export const MCP_SERVER_VERSION = "1.9.4";
+export const MCP_TOOLSET_VERSION = "2026.09.04.5";
+export const MCP_TOOLSET_CHANGED_AT = "2026-09-04T01:36:00Z";
export type McpToolsetInfo = {
serverVersion: string;
diff --git a/lib/mcp/ui-config.ts b/lib/mcp/ui-config.ts
index 8e07f130..9d7f5b2c 100644
--- a/lib/mcp/ui-config.ts
+++ b/lib/mcp/ui-config.ts
@@ -16,6 +16,8 @@ export function widgetResourceMeta(description: string, options?: { frameDomains
"openai/widgetDescription": description,
"openai/widgetPrefersBorder": true,
"openai/widgetDomain": MCP_UI_DOMAIN,
- "openai/widgetCSP": { connect_domains: connectDomains, resource_domains: resourceDomains, ...(frameDomains.length ? { frame_domains: frameDomains } : {}), redirect_domains: [MSO_ORIGIN] },
+ // Standard CSP fields live only in `ui.csp`. OpenAI documents this legacy
+ // key as still required for trusted `openExternal` redirect destinations.
+ "openai/widgetCSP": { redirect_domains: [MSO_ORIGIN] },
};
}
diff --git a/lib/mcp/ui-resources.test.ts b/lib/mcp/ui-resources.test.ts
index 1a655b7b..59eed6fb 100644
--- a/lib/mcp/ui-resources.test.ts
+++ b/lib/mcp/ui-resources.test.ts
@@ -4,7 +4,7 @@ vi.mock("server-only", () => ({}));
const { dispatch } = await import("./dispatch");
const { MCP_APP_MIME_TYPE, WORKFLOW_PROGRESS_URI, MSO_SURFACE_URI, readUiResource } = await import("./ui-resources");
-const { MCP_UI_DOMAIN } = await import("./ui-config");
+const { MCP_UI_DOMAIN, MSO_ORIGIN } = await import("./ui-config");
const { activeWorkflowForActor } = await import("@/lib/workflow");
const call = (name: string, args: Record<string, unknown> = {}) =>
@@ -73,13 +73,14 @@ describe("MCP Apps workflow progress UI", () => {
expect(content.text).toContain("Open directly");
expect(content.text).toContain("/assistant/mcp");
expect(content.text).toContain("Automatic open unavailable");
- expect(WORKFLOW_PROGRESS_URI).toContain("-v2.html");
+ expect(WORKFLOW_PROGRESS_URI).toContain("-v3.html");
expect(content.text).not.toContain("fetch(");
expect(content._meta).toMatchObject({
ui: { domain: MCP_UI_DOMAIN, prefersBorder: true },
"openai/widgetDomain": MCP_UI_DOMAIN,
"openai/widgetPrefersBorder": true,
});
+ expect(content._meta["openai/widgetCSP"]).toEqual({ redirect_domains: [MSO_ORIGIN] });
});
it("returns only redacted structured workflow state and keeps polling out of workflow memory", async () => {
@@ -162,12 +163,10 @@ describe("MCP Apps workflow progress UI", () => {
resourceDomains: [],
frameDomains: ["https://game.rahmanef.com"],
});
- expect(content._meta["openai/widgetCSP"]).toMatchObject({
- connect_domains: [],
- resource_domains: [],
- frame_domains: ["https://game.rahmanef.com"],
+ expect(content._meta["openai/widgetCSP"]).toEqual({
+ redirect_domains: [MSO_ORIGIN],
});
- expect(MSO_SURFACE_URI).toContain("surface-v4.html");
+ expect(MSO_SURFACE_URI).toContain("surface-v5.html");
});
it("keeps one dedicated progress card and one universal general-purpose Surface", () => {
diff --git a/lib/mcp/ui-resources.ts b/lib/mcp/ui-resources.ts
index 5825e32b..858db79c 100644
--- a/lib/mcp/ui-resources.ts
+++ b/lib/mcp/ui-resources.ts
@@ -2,7 +2,7 @@ import { MSO_SURFACE_RESOURCE } from "./ui-surface";
import { widgetResourceMeta } from "./ui-config";
import { OPEN_IN_MSO_SCRIPT, openInMsoControls } from "./ui-navigation";
export { MSO_SURFACE_URI } from "./ui-surface";
-export const WORKFLOW_PROGRESS_URI = "ui://mso/workflow-progress-v2.html";
+export const WORKFLOW_PROGRESS_URI = "ui://mso/workflow-progress-v3.html";
export const MCP_APP_MIME_TYPE = "text/html;profile=mcp-app";
const workflowProgressHtml = String.raw`<main class="mso-workflow" aria-live="polite">
diff --git a/lib/mcp/ui-surface.ts b/lib/mcp/ui-surface.ts
index 797fe557..a9ca81f0 100644
--- a/lib/mcp/ui-surface.ts
+++ b/lib/mcp/ui-surface.ts
@@ -4,7 +4,7 @@ import { surfaceFrameDomains } from "./surface-catalog";
import { MSO_SURFACE_SCRIPT } from "./ui-surface-script";
import { MSO_SURFACE_STYLE } from "./ui-surface-style";
-export const MSO_SURFACE_URI = "ui://mso/surface-v4.html";
+export const MSO_SURFACE_URI = "ui://mso/surface-v5.html";
const MIME = "text/html;profile=mcp-app";
const html = String.raw`<main class="surface" aria-label="MSO Surface">