feat: route coverage, docs, and completion for agent authoring story

docs/current_roadmap.md

@@ -171,6 +171,12 @@ Implementation order:
     rehearsal timing.
 24. Add a static slide/appendix shell only after presentation mode is clear.
     Astro remains an option, not the default next surface.
+25. Completed: presentation agent authoring story creates a canonical prepared
+    authoring recording, Scene 8 as an authentic full-screen conversation,
+    Scene 9 as a 5-phase lifecycle with trace panel, and route-level coverage
+    that never calls workflow authoring RPC operations during Scene 9
+    navigation. Implementation:
+    [`presentation agent authoring story`](historical/superpowers/plans/2026-07-11-presentation-agent-authoring-story.md).
 
 Presentation visual audit, July 10:
 
@@ -190,44 +196,12 @@ Presentation visual audit, July 10:
 - Completed: Scenes 13 and 14 now close with a bounded evaluation board,
   contribution boundary/future-work map, and canonical defense-question index.
 
-## Planned: Presentation Recomposition And Authoring Story
+## Completed: Presentation Recomposition And Authoring Story
 
-The presentation needs a deck-wide compositional pass, not more isolated
+Scenes 8 and 9 are now implemented as authentic external-agent handoff and
-opacity tweaks. A scene should establish an overview, then give one object the
+factual prepared-authoring proof. The canonical recording, full-screen Scene 8,
-stage while supporting evidence moves aside, exits, or enters as needed. A
+Scene 9 lifecycle with trace panel, and route coverage are all in place. Plan:
-focused object may expand into an inspectable graph, command/result surface, or
+[`presentation agent authoring story`](historical/superpowers/plans/2026-07-11-presentation-agent-authoring-story.md).
-discussion point before the next object takes focus. Comparisons remain visible
-when comparison is the point; unrelated support does not remain on stage by
-default.
-
-- Establish a reusable overview -> focus -> transition grammar across the
-  presentation. Motion should communicate spatial ownership, not merely dim
-  inactive elements. Reserve cyan workflow evidence for the demo range; use
-  editorial ink, paper, and amber outside it.
-- Recompose Scenes 1 through 7 around that grammar. Scene 2 should sequence
-  one-off chat/tool actions before durable automation. Scene 3 should stage
-  the left and right positioning panes before the substrate resolves between
-  them. Scene 6 should let architecture blocks expand into child figures.
-  Scene 7 should begin with the five-stage authoring overview, then focus each
-  stage with concrete command, result, diagnostic, or graph evidence.
-- Make Scene 8 a full-screen assistant-ui `Thread` that establishes the
-  external planner interface with no workflow chrome competing for attention.
-- Make Scene 9 the deterministic authoring proof. It uses a collapsible
-  assistant-ui `AssistantModal` opened by an `Agent trace` trigger over the
-  workflow canvas. Each authoring phase contains multiple literal `wf` tool
-  calls, then updates the main lifecycle/graph projection:
-  1. discover sources, capabilities, and schemas;
-  2. create the draft, steps, and outcome routes;
-  3. bind, validate, diagnose, and repair;
-  4. compile and inspect the immutable artifact;
-  5. create and validate the deployment.
-  Scene 10 begins only after that prepared deployment is legible.
-- Treat broken trace-frame display as a blocking product defect before the
-  run/evidence arc is redesigned around trace proof. Diagnose it separately
-  with a regression test and use factual run/trace data from the console path.
-- Recut crowded multi-beat scenes, including Scenes 13 and 14, so their beats
-  use distinct dominant artifacts instead of permanently rendering a complete
-  board or map with only minor emphasis shifts.
 
 Recommended next visual slices:
 
@@ -252,11 +226,15 @@ Recommended next visual slices:
    space. Implementation:
    [`guided proof scene composition cleanup`](historical/superpowers/plans/2026-07-10-guided-proof-scene-composition-cleanup.md).
 5. Completed: Evidence and closing visuals make Scenes 13 and 14 readable as a
-   defense artifact: bounded evaluation board, claim boundaries, future-work
+    defense artifact: bounded evaluation board, claim boundaries, future-work
-   map, and canonical examiner-question index. Design:
+    map, and canonical examiner-question index. Design:
-   [`presentation evaluation and closing`](superpowers/specs/2026-07-10-presentation-evaluation-closing-design.md).
+    [`presentation evaluation and closing`](superpowers/specs/2026-07-10-presentation-evaluation-closing-design.md).
-   Implementation:
+    Implementation:
-   [`presentation evaluation and closing plan`](historical/superpowers/plans/2026-07-10-presentation-evaluation-closing.md).
+    [`presentation evaluation and closing plan`](historical/superpowers/plans/2026-07-10-presentation-evaluation-closing.md).
+6. Completed: Presentation agent authoring story rebuilds Scenes 8 and 9 as
+    authentic external-agent handoff and factual prepared-authoring proof.
+    Implementation:
+    [`presentation agent authoring story plan`](historical/superpowers/plans/2026-07-11-presentation-agent-authoring-story.md).
 
 Presentation wishlist / defense readiness:
 

web/README.md

@@ -177,11 +177,14 @@ The final presentation beats frame the evaluation as bounded evidence, make
 claim boundaries and future work explicit, and end on the canonical defense
 discussion index rather than a benchmark or generic conclusion.
 
-Scenes 8 through 12 use the canonical replay as their only execution evidence.
+Scenes 8 and 9 use the canonical prepared-authoring recording as their only
-The handoff expands an interpreted run operation into the center stage, then
+execution evidence. Scene 8 is a full-screen conversation that establishes the
-keeps one workflow graph mounted while execution reaches the typed interrupt
+external agent handoff; Scene 9 breaks the prepared authoring into five phases
-and approval boundary. Raw protocol payloads are available through the evidence
+with an overlay trace panel. Neither scene calls workflow authoring RPC
-receipt and inspector.
+operations — they consume deterministic prepared data. Scenes 10 through 12 use
+the canonical replay to show run, interrupt, approval, resume, output, and
+trace. Raw protocol payloads are available through the evidence receipt and
+inspector.
 
 The presentation chat surface is source-owned and follows the AI Elements
 conversation/message/tool/prompt-action model. It currently renders the
@@ -195,6 +198,11 @@ The key defense states are directly addressable:
 - `/present#scene/workflow-demo/graph`
 - `/present#scene/workflow-demo/interrupt`
 - `/present#scene/interrupt-evidence/approval`
+- `/present#scene/agent-handoff/request` — Scene 8, prepared handoff request
+- `/present#scene/agent-handoff/handoff` — Scene 8, agent accepts
+- `/present#scene/prepared-lifecycle/discover` — Scene 9, discover phase
+- `/present#scene/prepared-lifecycle/draft` — Scene 9, draft phase
+- `/present#scene/prepared-lifecycle/deployment` — Scene 9, deployment phase
 
 Scene 12 is factual by design. It projects the reviewed/live run into visible
 workflow input, interrupt payload, resume decision, output, and trace facts.
@@ -307,10 +315,30 @@ The current prepared recipe can:
 This is intentionally not a general autonomous planner. A future server-side
 Vercel AI SDK driver can feed the same message-part interface.
 
-### Demo Climax (Scenes 9–12)
+### Authoring Story (Scenes 8 and 9)
 
-Scenes 9 through 12 are the demo climax. They keep a continuity rail visible while
+Scenes 8 and 9 are the prepared authoring story. They use deterministic data
-the prepared replay moves from agent handoff, to persisted workflow run, to
+from the committed `projectPreparedAuthoring()` recording and never call
-typed human interrupt, to resume/output/evidence. The rail and outcome panel are
+workflow authoring RPC operations.
+
+- **Scene 8 (Agent Handoff)**: a full-screen prepared conversation that
+  establishes the external agent interface. The `request` beat shows the
+  operator requesting a report workflow; the `handoff` beat reveals the full
+  completed conversation across all authoring phases.
+- **Scene 9 (Prepared Workflow Lifecycle)**: a five-phase lifecycle
+  (discover, draft, validate, artifact, deployment) with a compact phase rail
+  and one dominant command/result projection per beat. An overlay
+  `AuthoringTracePanel` surfaces the full agent trace for the active phase
+  without competing with the lifecycle projection.
+
+The authoring scenes consume deterministic prepared data and never call
+workflow authoring RPCs — only `workflow.health` infrastructure pings are
+permitted.
+
+### Demo Climax (Scenes 10–12)
+
+Scenes 10 through 12 are the demo climax. They keep a continuity rail visible
+while the prepared replay moves from persisted workflow run, to typed human
+interrupt, to resume/output/evidence. The rail and outcome panel are
 presentation-only projections over the committed replay; they do not add live
 backend dependencies.

web/apps/console/src/presentation/PresentationRoute.test.tsx

@@ -367,4 +367,49 @@ describe("PresentationRoute", () => {
     expect(await screen.findByRole("region", { name: /workflow trace frames/i })).toBeInTheDocument();
     expect(screen.queryByText("No trace frames captured.")).not.toBeInTheDocument();
   });
+
+  it("navigates to Scene 8 request beat via hash", async () => {
+    window.location.hash = "#scene/agent-handoff/request";
+    const { PresentationRoute } = await import("./PresentationRoute.js");
+    render(<PresentationRoute />);
+
+    expect(await screen.findByRole("heading", { name: "Agent Handoff" })).toBeInTheDocument();
+    expect(screen.getByText(/A thin agent interface receives the report request/)).toBeInTheDocument();
+  });
+
+  it("navigates to Scene 8 handoff beat via hash", async () => {
+    window.location.hash = "#scene/agent-handoff/handoff";
+    const { PresentationRoute } = await import("./PresentationRoute.js");
+    render(<PresentationRoute />);
+
+    expect(await screen.findByRole("heading", { name: "Agent Handoff" })).toBeInTheDocument();
+    expect(screen.getByText(/The interface delegates durable work to lda\.chat/)).toBeInTheDocument();
+  });
+
+  it.each([
+    "#scene/prepared-lifecycle/discover",
+    "#scene/prepared-lifecycle/draft",
+    "#scene/prepared-lifecycle/validate",
+    "#scene/prepared-lifecycle/artifact",
+    "#scene/prepared-lifecycle/deployment",
+  ])("navigates to Scene 9 beat %s", async (hash) => {
+    window.location.hash = hash;
+    const { PresentationRoute } = await import("./PresentationRoute.js");
+    render(<PresentationRoute />);
+
+    expect(await screen.findByLabelText("authoring phase rail")).toBeInTheDocument();
+  });
+
+  it("navigating Scene 9 beats does not call workflow authoring RPC operations", async () => {
+    const { callOperation } = await import("../connection/api.js");
+    const mockCallOp = vi.mocked(callOperation);
+    mockCallOp.mockClear();
+    window.location.hash = "#scene/prepared-lifecycle/draft";
+    const { PresentationRoute } = await import("./PresentationRoute.js");
+    render(<PresentationRoute />);
+
+    expect(await screen.findByLabelText("authoring phase rail")).toBeInTheDocument();
+    const authoringCalls = mockCallOp.mock.calls.filter((c) => c[0] !== "workflow.health");
+    expect(authoringCalls).toHaveLength(0);
+  });
 });