feat: ai-chat reference project + MCP agent-chat tooling (5/5)

[ericallam]

Layer 5 of 5 in the chat.agent stack split

Top of the chat.agent stack: a full Next.js reference project that
exercises chat.agent end-to-end, plus the CLI MCP tools that drive
agent runs from Claude Code / Cursor / etc.

Targets feature/agent-view-dashboard — merge last

references/ai-chat

• Full Next.js app with prisma persistence, multi-chat sidebar,
  per-chat model picker, debug panel, tool examples, smoke tests
• Reference tools: getCurrentTime, searchHackerNews,
  createGithubIssue, PR review helpers, code sandbox
• chat-client-test orchestrator for concurrent-send stress
• references/hello-world chatAgent + triggerAndSubscribe examples

CLI MCP tooling for chat.agent

• mcp/tools/agentChat.ts (start_agent_chat, send_agent_message,
  close_agent_chat)
• mcp/tools/agents.ts + tasks.ts (list agents, agent run details)
• dev-run-controller OOM kill + taskRunProcessPool tweaks
• dev/managed entry-point hooks for skills bundling
• buildWorker + bundleSkills (agent skills support)

Includes ai-tool-helpers + mcp-agent-chat-sessions changesets, plus
the streamdown@2 patch and pnpm-lock reconciliation.

Stack

• L1 → main: Sessions primitive
• L2 → L1: chat.agent runtime
• L3 → L2: browser chat client + transport
• L4 → L3: agent-view dashboard
• L5 → L4 (this PR)

[changeset-bot]

🦋 Changeset detected

Latest commit: 5baac29

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 30 packages

Name	Type
@trigger.dev/sdk	Patch
trigger.dev	Patch
@trigger.dev/python	Patch
@internal/sdk-compat-tests	Patch
references-ai-chat	Patch
d3-chat	Patch
references-d3-openai-agents	Patch
references-nextjs-realtime	Patch
references-realtime-hooks-test	Patch
references-realtime-streams	Patch
references-telemetry	Patch
@trigger.dev/build	Patch
@trigger.dev/core	Patch
@trigger.dev/react-hooks	Patch
@trigger.dev/redis-worker	Patch
@trigger.dev/rsc	Patch
@trigger.dev/schema-to-json	Patch
@trigger.dev/database	Patch
@trigger.dev/otlp-importer	Patch
@internal/cache	Patch
@internal/clickhouse	Patch
@internal/llm-model-catalog	Patch
@internal/redis	Patch
@internal/replication	Patch
@internal/run-engine	Patch
@internal/schedule-engine	Patch
@internal/testcontainers	Patch
@internal/tracing	Patch
@internal/tsql	Patch
@internal/zod-worker	Patch

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Pro

Run ID: 174b2c72-b6ac-4b94-b28a-55e87ba9cbff

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feature/ai-chat-reference-and-cli

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[devin-ai-integration]

Devin Review found 2 potential issues.

View 4 additional findings in Devin Review.

[devin-ai-integration]

🔴 Resetting lastEventId to undefined on upgrade causes SSE replay of entire session history

In collectAgentResponse, when a trigger:upgrade-required chunk is received, session.lastEventId is set to undefined (line 390) before recursively calling collectAgentResponse. The recursive call creates a new SSEStreamSubscription using session.lastEventId (now undefined) as the lastEventId option (agentChat.ts:338), which means the new subscription replays the entire session .out stream from the very beginning.

Since session .out is a durable stream containing all historical chunks across runs, the replayed events will include old turns' trigger:turn-complete chunks. The first historical trigger:turn-complete hit (line 366-368) immediately breaks the collection loop, causing the function to return with empty/partial text from a previous turn instead of the new run's response.

Expected fix

Keep session.lastEventId as-is (pointing to the trigger:upgrade-required chunk's SSE id) so the recursive subscription resumes right after the upgrade marker, where the new run's output will appear.

Suggested change

	session.lastEventId = undefined;
	reader.releaseLock();

---

Was this helpful? React with 👍 or 👎 to provide feedback.

[devin-ai-integration]

🚩 managed-index-controller adds prompts but not skills to worker registration payload

Both dev-index-worker.ts:187 and managed-index-worker.ts:183 now emit skills: resourceCatalog.listSkillManifests() in their worker manifests. The managed-index-controller.ts forwards tasks, prompts (newly added at line 107), and queues to the server — but does NOT forward skills. If skills need to be registered server-side (for dashboard display, deploy metadata, etc.), this omission would silently drop them in the managed (production) indexing path. Could be intentional if skills are only used at runtime and don't need server-side registration, but it's asymmetric with the prompts addition in the same PR.

---

Was this helpful? React with 👍 or 👎 to provide feedback.