Generate a structured paper outline from review conclusions and experiment results. Use when user says \"写大纲\", \"paper outline\", \"plan the paper\", \"论文规划\", or wants to create a paper plan before writing.

Set up TypeScript LangChain 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.

Add tools to your agent and grant required permissions in databricks.yml. Use when: (1) Adding MCP servers, Genie spaces, vector search, or UC functions to agent, (2) Permission errors at runtime, (3) User says 'add tool', 'connect to', 'grant permission', (4) Configuring databricks.yml resources.

Manage Lakebase Postgres Autoscaling projects, branches, and endpoints via Databricks CLI. Use when asked to create, configure, or manage Lakebase Postgres databases, projects, branches, computes, or endpoints.

Deploy applications and websites to Vercel. Use when the user requests deployment actions like "deploy my app", "deploy and give me the link", "push this live", or "create a preview deployment".

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.

Build apps on Databricks Apps platform. Use when asked to create dashboards, data apps, analytics tools, or visualizations. Invoke BEFORE starting implementation.

Monitor and manage Harness Delegates via MCP. List delegates and check health status, manage delegate registration tokens (create, revoke, delete), and find which delegates are associated with a token. Use when asked about delegate status, delegate health, delegate connectivity, delegate tokens, or troubleshooting delegate issues. Trigger phrases: delegate status, delegate health, list delegates, delegate token, delegate connectivity, delegate troubleshooting, delegate down.

Generate Harness Agent Template files for AI-powered automation agents. Produces metadata.json, pipeline.yaml (v1 syntax), and wiki.MD files. Agents automate tasks like code review, security scanning, test generation, and documentation. This is a YAML generation skill. Use when asked to create an agent template, build an AI agent, create an automation agent, generate agent pipeline files, or set up a Harness agent. Trigger phrases: create agent template, agent template, AI agent, automation agent, build agent, code review agent, security scanner agent, Harness agent.

Generate Harness Service YAML for deployable workloads and create via MCP. Supports Kubernetes, Helm, ECS, Serverless, SSH, and WinRm deployment types with artifact sources from Docker Hub, ECR, GCR, ACR, Nexus, and S3. Use when asked to create a service, define a Kubernetes service, set up a Helm chart deployment, configure an ECS service, or define what gets deployed. Trigger phrases: create service, service definition, Kubernetes service, Helm service, ECS service, deployment service, artifact source.

Generate Harness Environment YAML for deployment targets and create via MCP. Supports PreProduction and Production types with environment variables, manifest overrides, and multi-environment setup (dev, staging, prod). Use when asked to create an environment, set up staging, configure production, define deployment targets, or manage environment overrides. Trigger phrases: create environment, deployment environment, setup dev, setup staging, setup production, environment variables, environment overrides.

Use when the user says "get started with Cekura", "set up Cekura", "onboard to Cekura", "I'm new to Cekura", "help me set up my agent", "how do I use Cekura", "walk me through Cekura", "configure my project", "first time using Cekura", or needs guidance on initial platform setup. Covers two onboarding paths: **testing** (default — build evaluators and run simulated calls) and **observability** (ingest production call logs and evaluate them).

Implement dynamic form updates using Turbo Streams and Stimulus. Use when forms need to update fields based on user selections without full page reloads, such as cascading dropdowns, conditional fields, or dynamic option lists.

Implement Rails nested attributes with dynamic add/remove functionality using Turbo Streams and Simple Form. Use when building forms where users need to manage multiple child records (has_many associations), add/remove nested items without page refresh, or create bulk records inline.

Use when starting a new initiative spike, investigating technical feasibility, assessing impact across services, or writing an engineering discovery document — produces a structured spike doc with diagrams and a private notes file

Deploy and run ML experiments on local or remote GPU servers. Use when user says "run experiment", "deploy to server", "跑实验", or needs to launch training jobs.

Get a deep critical review of research from GPT using a secondary Codex agent. Use when user says "review my research", "help me review", "get external review", or wants critical feedback on research ideas, papers, or experimental results.

Use when writing or reviewing Java code in this repo that calls Deepgram Speech-to-Text v1 (`/v1/listen`) for prerecorded or live transcription. Covers `client.listen().v1().media().transcribeUrl` / `transcribeFile` (REST) and `client.listen().v1().v1WebSocket()` (WebSocket). Use `deepgram-java-audio-intelligence` for analytics overlays, `deepgram-java-conversational-stt` for Flux `/v2/listen`, and `deepgram-java-voice-agent` for full-duplex assistants. Triggers include "transcribe", "speech to text", "STT", "listen v1", "nova-3", "live transcription", and "websocket transcription".

Use when writing or reviewing JavaScript/TypeScript in this repo that builds an interactive voice agent via `agent.deepgram.com/v1/agent/converse`. Covers `client.agent.v1.createConnection()` / `connect()`, `sendSettings`, `sendMedia`, runtime updates, event handling, and function-call responses. Use `deepgram-js-text-to-speech` for one-way synthesis, `deepgram-js-speech-to-text` or `deepgram-js-conversational-stt` for transcription only, and `deepgram-js-management-api` for project/model admin rather than live agent runtime. Triggers include "voice agent", "agent converse", "full duplex", "barge-in", "function calling", and "agent.v1".