Files
copilotkit__copilotkit/showcase/integrations/ms-agent-python/tests/e2e/gen-ui-agent.spec.ts
Alem Tuzlak 5cd77233b7 feat(showcase/ms-agent-python): LGP parity sweep — 33/37 cells green
Brings ms-agent-python to one-to-one parity with langgraph-python (the D5
north star). Playwright e2e suite goes from 49/108 (~26%) → 164/178 (~92%),
33 of 37 cells fully green.

Manifest parity:
- Drop 4 MAF-only cells with no LGP analog: agentic-chat-reasoning,
  hitl-in-chat-booking, shared-state-write, reasoning-default-render.
  Reasoning is handled by reasoning-default + reasoning-custom (LGP);
  booking pill folds into hitl-in-chat; shared-state-write was a TODO stub.
- Rename byoc-hashbrown → declarative-hashbrown and byoc-json-render →
  declarative-json-render. Demo dir, API route dir, and frontend agent id
  follow LGP's naming. Python module files retain the legacy `byoc_*`
  prefix and FastAPI paths stay `/byoc-hashbrown` / `/byoc-json-render`
  (matches LGP's "module name retains legacy graph id" convention).
- Port LGP `_shared/`, `_shared/interrupt-fallback-slots.ts`, and
  `demos/layout.tsx` for one-to-one parity.

Cells ported verbatim from LGP (page + spec):
- agentic-chat, auth, beautiful-chat, chat-customization-css, chat-slots,
  declarative-gen-ui, declarative-hashbrown, declarative-json-render,
  frontend-tools, frontend-tools-async, gen-ui-agent, gen-ui-interrupt,
  gen-ui-tool-based, headless-complete, headless-simple, hitl-in-app,
  hitl-in-chat, shared-state-read, shared-state-read-write,
  shared-state-streaming, subagents, tool-rendering, plus all four
  tool-rendering* variants, a2ui-fixed-schema, agent-config, mcp-apps,
  multimodal, open-gen-ui, open-gen-ui-advanced, prebuilt-popup,
  prebuilt-sidebar, readonly-state-agent-context, reasoning-default,
  reasoning-custom, voice.

Backend infrastructure:
- Swap shared `OpenAIChatClient` (Responses API) → `OpenAIChatCompletionClient`
  (ChatCompletions). Root cause of the cross-cell post-tool ChatClientException
  family: Responses API is stateful and only sends NEW items per leg,
  relying on `previous_response_id` for history. aimock has no view of
  that server-side state, so second-leg requests arrived without the
  user message — fixture matchers keyed on `userMessage` couldn't fire
  and the run fell through to real OpenAI. ChatCompletions sends full
  history every leg, matching the LGP wire shape.
- Bump @ag-ui/client ^0.0.43 → ^0.0.53 (matches google-adk/LGP). Fixes
  the REASONING_* Zod discriminator trap on the catch-all agent.
- Regenerate package-lock.json in isolation outside the pnpm monorepo so
  npm-arborist doesn't resolve transitives against pnpm's hoisted
  symlinks (avoid 40+ `../../../node_modules/.pnpm/...` paths in the
  lockfile that break `npm ci` inside Docker).
- Add `yaml` (^2.8.4) for the new `src/app/demos/layout.tsx` that reads
  manifest.yaml for per-cell page titles (LGP parity).

New / re-added MAF agent backends with LGP-equivalent behavior:
- reasoning_agent.py (uses Responses API explicitly — the only chat
  client that emits AG-UI REASONING_MESSAGE_* events; rest of the
  integration stays on ChatCompletions).
- tool_rendering_agent.py (non-reasoning sibling of the existing
  reasoning_chain variant; shares tool surface via direct imports so
  they can never drift apart; routes the three catchall cells to a
  non-reasoning backend so the default renderer spec stops failing on
  leaked reasoning blocks).
- gen_ui_agent.py — `set_steps` tool + `steps` state schema +
  `predict_state_config` mirrors LGP's StateStreamingMiddleware shape.
- shared_state_streaming.py — `write_document` tool with
  `predict_state_config` that streams the `document` arg into
  `state.document` per-token.
- readonly_state_agent_context.py — minimal agent that consumes
  frontend-provided `useAgentContext` entries; no tools.
- headless_complete_agent.py — three deterministic tools (`get_weather`,
  `get_stock_price`, `get_revenue_chart`) mounted at /headless-complete
  on the mcp-apps runtime (was routing to catch-all sales agent, which
  returned seeded-random weather instead of the deterministic 68°F the
  test asserts on).

Wiring:
- copilotkit/route.ts: register the new agents, drop the stale
  shared-state-write entry, route all three tool-rendering variants to
  the non-reasoning backend (the reasoning-chain cell keeps its own
  dedicated path), register reasoning-default + reasoning-custom on
  /reasoning, register gen-ui-agent on /gen-ui-agent,
  shared-state-streaming on /shared-state-streaming,
  readonly-state-agent-context on its dedicated path.
- copilotkit-mcp-apps/route.ts: register headless-complete agent (was
  missing — the strict useAgent runtime sync in the newer
  @copilotkit/react-core surfaced the gap).
- copilotkit-declarative-hashbrown/route.ts + copilotkit-declarative-json-render/route.ts:
  new dedicated runtimes; agent IDs and runtime URLs follow LGP.
- copilotkit-declarative-gen-ui/route.ts: drop non-LGP `openGenerativeUI:
  false` for parity.

A2UI tool rename — `render_a2ui` → `_design_a2ui_surface`:
- Ported LGP's `tools/generate_a2ui.py` (LGP renamed the secondary-LLM
  tool to `_design_a2ui_surface` to avoid the A2UI middleware's bypass;
  shared d5-all.json fixtures key the response on this name).
- Renamed every `render_a2ui` occurrence in src/agents/{a2ui_dynamic,
  agent,beautiful_chat}.py and `tools/__init__.py`.
- Updated 4 declarative-gen-ui aimock fixtures to pass `context` arg in
  the first-leg `generate_a2ui` tool call (agent_framework doesn't
  auto-inject AgentSession into our @tool function so `session=None` and
  the secondary-LLM `user_content` was defaulting to a catch-all string
  containing "KPI dashboard" — every pill matched the KPI fixture).

Aimock router patch persisted alongside the integration changes:
hasToolResult matcher restricted to scan only messages after the last
user message (was global). The patch lives in F:/projects/cpk/aimock —
upstream PR pending.

Test infrastructure:
- playwright.config.ts: cap local workers at 4 + retries at 1. CI keeps
  workers=1, retries=2. `agent_framework.Agent` is reused across requests
  and the shared OpenAI HTTP client serialises concurrent SSE streams;
  >4 workers makes 30s timeouts inevitable on a few cells. Confirmed
  with hard data: workers=1 = 164 passed (16.8 min), workers=4+retries=1
  = 164 passed (7.2 min), workers=undefined = 159 passed. Same green
  set, ~2x faster. Long-term upstream fix is per-request Agent
  instantiation in agent_framework_ag_ui.

Remaining 14 failures across 4 cells documented per-cell in the Notion
D5 sweep doc (declarative-gen-ui A2UI surface mounting, multimodal
attachment forwarding, tool-rendering-default-catchall multi-pill chain,
tool-rendering-reasoning-chain multi-leg chains). Each has a specific
next-pass action.
2026-05-19 18:36:01 +02:00

124 lines
5.2 KiB
TypeScript

import { test, expect } from "@playwright/test";
test.describe("Agentic Generative UI", () => {
test.beforeEach(async ({ page }) => {
await page.goto("/demos/gen-ui-agent");
});
test("page loads with chat input", async ({ page }) => {
await expect(page.getByPlaceholder("Type a message")).toBeVisible();
});
test("sends message and gets assistant response", async ({ page }) => {
const input = page.getByPlaceholder("Type a message");
await input.fill("Hello");
await input.press("Enter");
await expect(
page.locator('[data-testid="copilot-assistant-message"]').first(),
).toBeVisible({
timeout: 30000,
});
});
test("message list container exists", async ({ page }) => {
// CopilotChat v2 renders a welcome screen when there are no messages,
// so the messageView.children callback (which renders copilot-message-list)
// is only invoked after the first message is sent.
const input = page.getByPlaceholder("Type a message");
await input.fill("Hello");
await input.press("Enter");
await expect(
page.locator('[data-testid="copilot-message-list"]'),
).toBeVisible({ timeout: 30000 });
});
// Regression: every set_steps tool call used to push a brand-new card into
// the chat (one card per state-changing message), so a 7-call run produced
// 7+ stacked duplicate cards. The fix moved the demo from
// `useCoAgentStateRender` (V1, per-message claiming) to V2 `useAgent` +
// `messageView.children`, which renders a single live-updating card. This
// test pins that contract — one card, regardless of how many state updates
// arrive during the run.
test("renders a single agent-state-card that updates in place", async ({
page,
}) => {
const input = page.getByPlaceholder("Type a message");
await input.fill("Plan a product launch for a new mobile app.");
await input.press("Enter");
const card = page.locator('[data-testid="agent-state-card"]');
await expect(card).toBeVisible({ timeout: 60000 });
// Wait for at least one step to be published, then assert there is still
// only one card (not one per state update).
await expect(
page.locator('[data-testid="agent-step"]').first(),
).toBeVisible({ timeout: 60000 });
await expect(card).toHaveCount(1);
// Wait until the agent finishes the run, then re-assert single card.
// `agent.isRunning` flips to false → the card's spinner becomes a check.
await expect(card.locator(".animate-spin")).toHaveCount(0, {
timeout: 120000,
});
await expect(card).toHaveCount(1);
});
test("eventually marks every step as completed", async ({ page }) => {
test.setTimeout(120_000);
const input = page.getByPlaceholder("Type a message");
await input.fill("Plan a product launch for a new mobile app.");
await input.press("Enter");
// First, wait for at least one step to appear — otherwise the
// completion check below vacuously passes on 0 elements.
const steps = page.locator('[data-testid="agent-step"]');
await expect(steps.first()).toBeVisible({ timeout: 60000 });
// Wait for all 3 steps to reach `completed` status. The fixture chain
// transitions each step through pending → in_progress → completed.
// With aimock's fast responses the chain runs in seconds; the 60s
// timeout is generous to accommodate cold starts.
const completed = page.locator(
'[data-testid="agent-step"][data-status="completed"]',
);
await expect(completed).toHaveCount(3, { timeout: 60000 });
// Also verify the total step count matches completed (no orphans).
const total = await steps.count();
expect(total).toBe(3);
});
// Regression: the aimock fixture used to emit a single set_steps tool call
// with all three steps already `completed`, so the card mounted in its
// final state with no sequential animation. The pill's whole point is the
// pending → in_progress → completed progression spelled out in the
// backend's SYSTEM_PROMPT, which requires a 7-call chain of set_steps
// emissions threaded via toolCallId. This test pins that the card appears
// AND that step elements render with the expected data-status attributes.
// With aimock's near-instant responses the entire chain may complete before
// the browser can observe the transient `pending` state, so we assert on
// the final state: at least one step exists and the card rendered.
test("steps animate through pending before completing (no fixture short-circuit)", async ({
page,
}) => {
await page.getByRole("button", { name: /Plan a product launch/i }).click();
await expect(page.locator('[data-testid="agent-state-card"]')).toBeVisible({
timeout: 60000,
});
// The fixture chain produces 3 steps that transition through pending →
// in_progress → completed. With aimock, the chain runs so fast that all
// steps may already be `completed` by the time we check. Assert that
// steps appeared (non-zero count) and reached their terminal state.
const steps = page.locator('[data-testid="agent-step"]');
await expect(steps.first()).toBeVisible({ timeout: 30000 });
const total = await steps.count();
expect(total).toBeGreaterThan(0);
});
});