Skip to content

datarobot-oss/datarobot-agent-skills / skills/datarobot-agent-assist/agent-assist-build

datarobot-agent-assist-build

Flagged

Use when the user wants to design, build, code, or deploy an AI agent on DataRobot; mentions agent_spec.yaml, dress rehearsal, the DataRobot agent template, LangGraph, CrewAI, LlamaIndex, NAT, Base agents, MCP servers, backend APIs, custom frontends, or the DataRobot CLI.

references/agent-spec-examples.md

View on GitHub
# agent_spec.yaml Examples

Field definitions and structure: [agent-spec-schema.md](agent-spec-schema.md).

## Example 1: Weather Agent (Simple)

```yaml
model: datarobot/anthropic/claude-haiku-4-5-20251001
system_prompt: You are a helpful weather assistant that provides current weather conditions
  for any location worldwide. When a user asks about weather, search for current weather
  information and present the results in a clear, conversational way. Focus on current
  conditions like temperature, weather description, and other relevant details.
tools:
  - function_name: search_weather
    inputs:
      - arg_name: location
        type: str
    out:
      - arg_name: search_results
        type: str
    auth_spec:
      service_name: Perplexity API
      auth_method: api_key
examples:
  - What's the weather like in New York?
  - How's the weather in Tokyo today?
  - Current conditions in London
frontend:
  type: chat
```

---

## Example 2: Multi-tool Research Agent (with Auth)

```yaml
model: datarobot/anthropic/claude-sonnet-4-5-20250929
system_prompt: You are a research assistant that helps users find and summarize
  information from internal documents and the web. Always cite your sources.
tools:
  - function_name: search_internal_docs
    inputs:
      - arg_name: query
        type: str
      - arg_name: top_k
        type: int
    out:
      - arg_name: documents
        type: list
        object_schema: "list of {title: str, content: str, url: str}"
    auth_spec:
      service_name: Internal Knowledge Base API
      auth_method: bearer_token
  - function_name: web_search
    inputs:
      - arg_name: query
        type: str
    out:
      - arg_name: results
        type: str
examples:
  - Find recent papers on LLM hallucination
  - What does our internal policy say about data retention?
  - Summarize the latest news on AI regulation
frontend:
  type: chat
```

---

## Example 3: Multi-page Dashboard Agent

```yaml
model: datarobot/vertex_ai/gemini-2.5-pro
system_prompt: You are a sales analytics assistant. Help users understand their
  pipeline, forecast revenue, and identify at-risk deals. Always ground your
  answers in the data returned by tools.
tools:
  - function_name: get_pipeline_data
    inputs:
      - arg_name: date_range
        type: str
        object_schema: "ISO 8601 date range, e.g. '2024-01-01/2024-03-31'"
    out:
      - arg_name: deals
        type: list
        object_schema: "list of {deal_id: str, stage: str, value: float, close_date: str, owner: str}"
    auth_spec:
      service_name: Salesforce CRM
      auth_method: oauth2
  - function_name: forecast_revenue
    inputs:
      - arg_name: period
        type: str
    out:
      - arg_name: forecast
        type: dict
        object_schema: "{expected: float, low: float, high: float, confidence: float}"
examples:
  - What's our Q2 pipeline look like?
  - Which deals are at risk of slipping?
  - Forecast revenue for next quarter
frontend:
  type: multi-page
  pages:
    - "Pipeline Overview - shows all deals by stage with filtering"
    - "Revenue Forecast - charts expected vs actual with confidence bands"
    - "At-Risk Deals - highlights deals with low probability or stalled activity"
  requirements: "Charts should use a dark theme. Pipeline table must be sortable and filterable by owner and stage."
```

---

## Example 4: On-prem Agent (DataRobot-deployed LLM)

Use when the instance has no LLM Gateway catalog and the agent routes to an
existing text-generation deployment instead.

```yaml
model: datarobot/datarobot-deployed-llm
llm_deployment_id: "6a43eb5f10dbecadbebc5b2b"
system_prompt: You are an internal documentation assistant. Answer questions using
  the search_docs tool and cite the document titles you used.
tools:
  - function_name: search_docs
    inputs:
      - arg_name: query
        type: str
    out:
      - arg_name: matches
        type: list
        object_schema: "list of {title: str, excerpt: str}"
examples:
  - How do I configure runtime parameters?
  - Where is the deployment checklist documented?
frontend:
  type: chat
```

---

## Auth Method Reference

| `auth_method` | When to use |
|---|---|
| `api_key` | Static key passed in header or query param (e.g. OpenAI, Perplexity, SendGrid) |
| `oauth2` | User-delegated access with token refresh (e.g. Salesforce, Google, GitHub) |
| `basic_auth` | Username + password (e.g. legacy internal APIs) |
| `bearer_token` | Static bearer token (e.g. DataRobot API, internal services) |
| `service_account` | Non-human identity with key file or IAM role (e.g. GCP, AWS) |
| `other` | Anything that doesn't fit the above |

## Frontend Type Reference

| `type` | When to use |
|---|---|
| `chat` | Default — single chat window, no additional pages needed |
| `multi-page` | Multiple distinct pages (dashboard tabs, admin panel, etc.) |
| `custom` | Fully custom UI that requires bespoke layout beyond pages |
writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deploymentStars: 323kUpdated Claude CodeCodexVerifiedInstall
xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release.Stars: 118kUpdated Claude CodeCodexVerifiedInstall