菜单
Set up the Databricks chatbot app for local development and deployment. Use when: (1) First time setup, (2) User says 'quickstart', 'set up', 'authenticate', or 'configure databricks', (3) No .env file exists, (4) User says 'enable feedback', 'feedback widget', or 'MLFLOW_EXPERIMENT_ID'.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Give an agent durable, cross-session long-term memory using Databricks MANAGED memory (the Unity Catalog memory-store REST APIs) as tools — governed by UC with no infra the customer needs to run. This works for either OpenAI Agents SDK or LangGraph templates. Use when: the agent should remember a user's (or a team/org's shared) preferences/facts/decisions across conversations; keywords 'long-term memory', 'managed memory', 'memory store', 'agentic memory'. This is separate from the self-hosted Lakebase memory solution with skills in (agent-openai-memory / agent-langgraph-memory).
Configure Lakebase for agent memory storage. Use when: (1) Adding memory capabilities to the agent, (2) 'Failed to connect to Lakebase' errors, (3) Permission errors on checkpoint/store tables, (4) User says 'lakebase', 'memory setup', or 'add memory'.
Set up Databricks agent development environment. Use when: (1) First time setup, (2) Configuring Databricks authentication, (3) User says 'quickstart', 'set up', 'authenticate', or 'configure databricks', (4) No .env file exists.
Set up Databricks agent development environment. Use when: (1) First time setup, (2) Configuring Databricks authentication, (3) User says 'quickstart', 'set up', 'authenticate', or 'configure databricks', (4) No .env file exists.
Set up Databricks agent development environment. Use when: (1) First time setup, (2) Configuring Databricks authentication, (3) User says 'quickstart', 'set up', 'authenticate', or 'configure databricks', (4) No .env file exists.
Replace the client-side agent loop with Databricks Supervisor API (hosted tools + client-side function tools). Use when: (1) User asks about Supervisor API, (2) User wants Databricks to run the agent loop server-side, (3) Connecting Genie spaces, UC functions, agent endpoints, or MCP servers as hosted tools, (4) Mixing client-side function tools (Python callables your app executes) with hosted tools.
| name | quickstart |
| description | Set up the Databricks chatbot app for local development and deployment. Use when: (1) First time setup, (2) User says 'quickstart', 'set up', 'authenticate', or 'configure databricks', (3) No .env file exists, (4) User says 'enable feedback', 'feedback widget', or 'MLFLOW_EXPERIMENT_ID'. |
Run the interactive setup script from the e2e-chatbot-app-next/ directory:
./scripts/quickstart.sh
The script walks through the following steps interactively:
databricks auth loginMLFLOW_EXPERIMENT_ID in .envdatabricks.ymlMLFLOW_EXPERIMENT_ID in app.yamldb-chatbot-dev-<username>, max 30 chars)databricks bundle deploy and starts the appIf you skipped feedback during quickstart, or ran the script before feedback support was added, enable it manually:
1. Find the experiment name for your endpoint:
npx tsx scripts/get-experiment-id.ts --endpoint <your-serving-endpoint-name>
For Agent Bricks (Knowledge Assistant / Multi-Agent Supervisor):
npx tsx scripts/get-experiment-id.ts --agent-brick <agent-brick-name>
2. Set MLFLOW_EXPERIMENT_ID in .env (for local dev):
MLFLOW_EXPERIMENT_ID=<experiment-name-from-step-1>
3. Configure databricks.yml — uncomment and fill in the experiment resource:
- name: experiment
description: "MLflow experiment for collecting user feedback"
experiment:
name: "<experiment-name-from-step-1>"
permission: CAN_EDIT
4. Configure app.yaml — uncomment the env var:
- name: MLFLOW_EXPERIMENT_ID
valueFrom: experiment
5. Redeploy:
databricks bundle deploy
databricks bundle run databricks_chatbot
| Issue | Solution |
|---|---|
| Feedback widget not showing after setup | Restart dev server; env vars are read at startup |
get-experiment-id.ts fails with auth error | Run databricks auth login first |
| No experiment found for endpoint | Only custom agents and Agent Bricks endpoints have linked experiments; Foundation Model endpoints do not support feedback |
| Feedback submission returns 403 | App service principal is missing CAN_EDIT on the experiment — check permission: CAN_EDIT in databricks.yml |
| "Instance name is not unique" on deploy | Run ./scripts/cleanup-database.sh to remove the old database instance |