mirror of
https://github.com/CopilotKit/CopilotKit.git
synced 2026-09-14 16:26:20 +08:00
Alem Tuzlak
23a3b24a01
Adds real working Shared State (Read+Write) and Sub-Agents demos to 15 showcase integrations, mirroring the canonical langgraph-python and google-adk reference implementations. Fills rows previously empty on the showcase coverage dashboard. Packages: ag2, agno, claude-sdk-python, claude-sdk-typescript, crewai-crews, langgraph-fastapi, langgraph-typescript, langroid, llamaindex, mastra, ms-agent-dotnet, ms-agent-python, pydantic-ai, spring-ai, strands. (built-in-agent landed independently on main as PR #4321 — its variant is canonical; this PR no longer touches it.) Per-package deliverables: framework-native backend agents (preferences-injection middleware/callback + set_notes tool; supervisor + 3 sub-agents wired as tools with running -> completed /failed delegation log); frontend page.tsx + preferences-card.tsx / notes-card.tsx for SSRW and delegation-log.tsx for subagents — wired to useAgent({ updates: [OnStateChanged] }); manifest entries; runtime route registration + per-package agent server config; real QA scripts. Includes targeted hardening fixes from a 7-agent code-review loop: - Sub-agent failure paths now correctly emit status: "failed" (previously hardcoded "completed" or unreachable in mastra/strands/langgraph-fastapi/langgraph-typescript/ag2) - Parallel-tool-call delegation race fixed in langgraph-fastapi (Annotated[list, add]) and langgraph-typescript (concat reducer) - Silent data loss eliminated in claude-sdk-python/claude-sdk-typescript/crewai-crews — empty JSON.parse catches now log + emit error events - ms-agent-dotnet set_notes writes to per-thread slot via AsyncLocal - mastra working-memory writes are deterministic via src/mastra/tools/working-memory.ts helper - spring-ai tool-call envelope ids match supervisor's tc.id() and AG-UI event ordering reordered; CopyOnWriteArrayList for parallel-call safety - Stack trace + raw error message leaks scrubbed across 8+ Next.js routes — log server-side with errorId + return generic envelope - Sub-agent calls no longer block event loops in ag2 (asyncio.to_thread), langroid (llm_response_async), pydantic-ai (async run + async tools) - langroid lru_cache cross-request contamination dropped - Numerous smaller items: claude-sdk-python invalid model id, Callable annotation, /health endpoint exposed; crewai-crews supervisor no longer resets delegations every turn; pydantic-ai snapshot uses model_dump() CI fixes folded in: - crewai-crews test_forwarded_props: extend the stubbed ag_ui_crewai.endpoint module to expose add_crewai_flow_fastapi_endpoint and add stub agents.shared_state_read_write / agents.subagents modules - generate-catalog test: bump crewai-crews wired-cell expectation 28 -> 30; replace hardcoded total-wired count with an invariant (wired + stub + unshipped = 737) plus a lower-bound floor - oxfmt run on the qa/shared-state-read-write.md files in mastra + spring-ai Rebased onto latest main (post showcase/packages -> showcase/integrations rename + post built-in-agent landing). Original blitz history preserved at the blitz-pre-rebase-snapshot tag. Known follow-ups (deferred to follow-up PR): - agno sync sub_agent.run() blocks event loop (perf only) - ms-agent-python asyncio thread-fallback fragility - llamaindex initial-state coercion when UI clears state - Manifest highlight audit (langgraph-typescript headless-complete, langgraph-fastapi byoc-* missing route.ts highlights) - agno hitl-in-chat declared in demos but not features; duplicate /demos/hitl-in-chat route - langgraph-typescript server.mjs graphSpec only registers 3 graphs vs 23 in langgraph.json (pre-existing) - mastra hitl legacy demo missing from features list - claude-sdk-python agents/agent.py line 474 also has the legacy claude-opus-4-5 default - PARITY_NOTES vs manifest mismatches for hitl-in-app across spring-ai/agno/ag2 (pre-existing) - spring-ai a2ui-fixed-schema missing from generative_ui list
3.9 KiB
3.9 KiB
QA: Sub-Agents — CrewAI (Crews)
Prerequisites
- Demo is deployed and accessible at
/demos/subagentson the dashboard host - Agent backend is healthy (
/api/health);OPENAI_API_KEYis set; the FastAPI agent server has the/subagentsFlow endpoint mounted (seesrc/agent_server.py)
Test Steps
1. Basic Functionality
- Navigate to
/demos/subagents; verify the page renders within 3s with the left-side delegation log panel and right-sideCopilotChatpane - Verify
data-testid="delegation-log"is visible with heading "Sub-agent delegations" - Verify
data-testid="delegation-count"reads "0 calls" before any task is sent - Verify the empty-state copy "Ask the supervisor to complete a task. Every sub-crew it kicks off will appear here." is visible
- Verify the chat input placeholder is "Give the supervisor a task..."
- Verify all 3 suggestion pills are visible: "Write a blog post", "Explain a topic", "Summarize a topic"
2. Feature-Specific Checks
Supervisor Delegates to Sub-Crews
- Click the "Write a blog post" suggestion (sends a message asking for a short blog post about cold exposure with research → write → critique)
- Within ~3s verify
data-testid="supervisor-running"("Supervisor running" pulsing badge) appears in the header while the run is in progress - Within 60s verify three
data-testid="delegation-entry"rows appear in order:- A "Research" badge (purple, 🔎) with
data-testid="delegation-status"readingcompletedand the result body containing 3-5 bullets prefixed with- - A "Writing" badge (green, ✍️) with status
completedand a single-paragraph result - A "Critique" badge (orange, 🧐) with status
completedand 2-3 bullet-point critiques
- A "Research" badge (purple, 🔎) with
- Verify
data-testid="delegation-count"reads "3 calls" (or the matching number) at the end of the run - Verify the supervisor's final assistant message in the chat is short and references that the work is done (it is told to keep its own messages short)
Live Status Transitions
- Click the "Explain a topic" suggestion
- Watch a delegation entry first appear with status
running(its result body shows "Sub-agent running...") before transitioning tocompletedonce the sub-crew kickoff returns — confirms the flow emits a STATE_SNAPSHOT both before and after eachCrew.kickoff(...)call - Verify the
data-testid="supervisor-running"badge disappears once the run finishes
Multiple Tasks in One Session
- After the first run completes, send "Now do the same for solar power adoption."
- Verify the delegations list is reset for the new turn (the flow's
supervise()step clearsstate.delegationsat the top of each turn) and grows again as the new sub-agents are called - Verify each sub-agent's result body is non-empty and corresponds to its role (research is bullets, writing is a paragraph, critique is bullets)
3. Error Handling
- Send an empty message; verify it is a no-op (no user bubble, no supervisor invocation)
- Send a deliberately ambiguous message like "Hi"; verify the supervisor either responds directly without delegating (delegation count stays at 0) or kicks off at most one sub-crew — both are acceptable, but the chat must produce a final assistant message and no entries should be stuck in
runningafter the run finishes - Verify DevTools -> Console shows no uncaught errors during any of the above
Expected Results
- Page loads within 3 seconds
- A typical research → write → critique run completes within ~60 seconds with 3 delegation entries, all
completed - Each delegation entry transitions through
running→completed(orfailedif a sub-crew errors) and never gets stuck inrunningafter the supervisor has finished - The delegation log resets at the start of every fresh user turn
- No UI layout breaks, no uncaught console errors