ChatDVN

<script src="https://dvntmprodsa.blob.core.windows.net/$web/Extensions/Insert_Code_Button.js?sp=r&st=2025-02-03T06:00:00Z&se=2027-02-04T06:00:00Z&spr=https&sv=2022-11-02&sr=b&sig=l4bCeNdC4lv5dF%2FmXDbrZQss%2FHHHIrMezXWmRKW7M1g%3D"></script>


<script src="https://dvntmprodsa.blob.core.windows.net/$web/Extensions/Quote_Selection.js?sp=r&st=2025-02-03T06:00:00Z&se=2027-02-04T06:00:00Z&spr=https&sv=2022-11-02&sr=b&sig=J0uH%2BSJxzNEYfXVY25a2JK1y7dvM3VddzcHVzBKKU8g%3D"></script>

Content-Security-Policy: frame-ancestors https://*.cloud.microsoft;
X-Frame-Options: ALLOW

ChatDVN v3 Dev

Azure

Executes a Snowflake SQL query and returns the results. Suggest using the Alation plugin to get more information about tables. Format as a table using markdown by default.

query_snowflake

Snowflake Query

search_alation

Maximum number of results to return in request

Index of first search result to return from full result set

Alation Search

search_alation_knowledge_articles

Alation Knowledge Articles

Searches Wikipedia for a given keyword and retrieves a specific section of the article content based on the provided section index.

retrieve_wikipedia_article_specific_section

The index of the section to retrieve from the article

Wikipedia

Generate images using DALL-E 3 based on image descriptions. Adhere to guidelines such as English translation, not depicting politicians or recent artists, and ensuring artist references are over 100 years old.

When creating prompts: substitute artist names with style-describing adjectives, include artistic era, mention primary medium. Image type, descent, gender are to be explicitly stated. Avoid offense and bias, conceal identities except gender and physique, retain text references, replace profession titles with universal terms.

Prompt changes aren't allowed unless they anonymise individuals. Replace creative professional/studio names with style descriptions. All prompts should detail every image part in objective terms within a detailed paragraph of over three sentences.

Important: after calling this function, the output images will be shown to the user automatically. You don't need to include the output image to the users again in your response to avoid duplicate images.

server_generate_image_via_dalle_3

The description of one (the first) image as defined in the guidelines.

Azure DALL-E 3

search_bing

Bing Search

search_bing_images

The number of image search results to return

The offset of the first image result to return

Bing Image Search

search_bing_news

The number of news search results to return

The offset of the first news result to return

Bing News Search

Query the Wolfram Alpha API and return the result

query_wolfram_alpha

The query string to search for in Wolfram Alpha

Wolfram Alpha

render_chart

Additional options for the Chart.js chart

Render Chart

Render an interactive canvas with HTML source to the user interface. The HTML source can include JavaScript and CSS to create interactive elements. This can be used to create custom user interfaces, games, demos, charts, and more. The canvas width is always 100% of the container width, and the height can be specified in pixels.

render_interactive_canvas

The height of the canvas in pixels. Default is 500.

Interactive Canvas

get_datetime

The IANA time zone name, e.g. America/Los_Angeles

DateTime

Query the OPIS Spot API for spot pricing data. The API contains spot pricing data for various oil and gas products.

query_opis_spot_pricing

The end date for the spot pricing data. In the format YYYY-MM-DD. (optional for date range selection)

The comma delimited list of OPIS product codes.

The start date for the spot pricing data. In the format YYYY-MM-DD. (optional for date range selection)

OPIS Spot Prices

Search Cognitive Cache, Devon's enterprise vector database, for relevant information. When the user asks you something, you can use this function to search for relevant information in Cognitive Cache. The search query should be related to the conversation.

query_cogcache_hr_docs

The phrase or sentence to search for in Cognitive Cache, Devon's enterprise vector database.

CogCache - HR

Get the chunked contents of a document by Id from Devon's Cognitive Cache service. You must know the Document Id before calling this function. If needed perform a Cognitive Cache search to discover a specific document's Id.

get_cognitive_cache_document_by_id

Cognitive Cache Get Document By Id

Perform a semantic search against the Devon Cognitive Cache document store and return a list of relevant documents (includes document id).

list_relevant_cognitive_cache_documents

The target security group to search, representing the data domain

Cognitive Cache List Relevant Documents

Gets a list of available captions for a youtube video.

get_youtube_caption_ids

Youtube Client Side Test

Generate a Cypher query from a user-provided input.

employee_search_generate_query

Devon Employee Search - Generate Query

Execute a Cypher query and return the result from the database.

employee_search_execute_query

Devon Employee Search - Execute Query

oauth_plugin_test

OAuth Graph API Test plugin

This renders the markdown content as an artifact in the user interface.

render_markdown_artifact

Render Markdown as Artifact

Perform a semantic search against the Devon Cognitive Cache document store for a single data domain.

search_cognitive_cache_data_domain

Cognitive Cache Search Data Domain

Perform a semantic search against the Devon Cognitive Cache document store.

search_cognitive_cache_personal

Cognitive Cache Search Personal Docs

Retrieves table metadata from Alation by table ID.

alation_table_by_id

Include published queries in the response

Alation Table Metadata By Id

Retrieves columns metadata from Alation by table ID.

alation_columns_by_table_id

Alation Columns Metadata By Table Id

Retrieves a user's email address from ChatDVN's API using the system-provided {USER_ID} parameter. This function automatically uses the built-in user context and requires no additional parameters from the LLM or the user. Returns the email address associated with the current user context.

get_user_email_by_id

Indicates if the email retrieval was successful

Get User Email By Id

This renders citations in an artifact to allow side by side viewing. This tool should be called in a message AFTER the users question has been answered. Make sure that all images use a width=100% so that they fit and resize with their container.

render_citations

View Citations

Generates an interactive Monaco Editor-based HTML page with syntax highlighting, a copy button, and a customizable title.

code_editor_artifact

The initial code content to be preloaded into the Monaco Editor.

The programming language mode for syntax highlighting. Supports multiple languages such as JavaScript, Python, and more.

The title of the code editor page, displayed in the browser tab.

Code Artifact

Execute a Python code snippet that evaluates to a readable value. Use this function to evaluate a Python expression. The Python will be executed on the user's local browser in a secure sandboxed context using Pyodide (WASM python running in browser).

execute_python

A valid Python code snippet to be executed in the Pyodide Python WASM runtime.

Python Interpreter

Retrieve weather data for a specified location. Users can customize which weather variables to include based on their preferences. If a state is provided, never use the abbreviation, instead always spell out the full name (i.e. Texas instead of TX).

get_weather

The number of results to return for location searches. Adjust if necessary to find the correct location.

determine whether to current weather data.

Determine whether to include daily weather data.

Determine whether to include hourly weather data.

The language for the location search. Use ISO 639-1 two-letter language codes.

Get Weather

Comma-separated list of current weather variables to fetch. Default is 'temperature_2m,relative_humidity_2m,is_day,precipitation,wind_speed_10m,weather_code'.

current_weather

Comma-separated list of hourly weather variables to fetch. Default is 'temperature_2m,relative_humidity_2m,precipitation_probability,wind_speed_10m,weather_code'.

hourly_weather

Comma-separated list of daily weather variables to fetch. Default is 'weather_code,temperature_2m_max,temperature_2m_min,precipitation_probability_max'.

daily_weather

The unit for temperature. Default is 'fahrenheit'.

temperature_unit

The timezone for the weather data. Default is 'auto'.

timezone

Search the DEBI Agent knowledge base for relevant information. When the user asks you something, you can use this function to search for relevant information in the DEBI Agent knowledge base. Don't use questions as search query, use keywords or phrases related to the conversation instead. 

query_debi_agent_knowledge_base

The keywords or topics to search for in the DEBI Agent knowledge base. Multiple keywords or topics are accepted (separated with commas).

DEBI Agent Knowledge Base

Calculate a math expression. For example, "2 + 2" or "2 * 2". The expression must be a valid JavaScript math expression.

get_calculation_result

A valid JavaScript math expression for the calculation.

Simple Calculator

Searches TwelveData for a given stock ticker symbol and returns meta and time series for the requested instrument. Metaobject consists of general information about the requested symbol. Time series is the array of objects ordered by time descending with Open, High, Low, Close prices. Non-currency instruments also include volume information. Only up to 30 values are returned per symbol.  Adjust the {sample_rate} accordingly to fulfill the request.  If 30 days of data are request, use a 1day sample_rate.  If 30 minutes of data are requested, use a 1min sample_rate. Supported intervals: 1min, 5min, 15min, 30min, 45min, 1h, 2h, 4h, 8h, 1day, 1week, 1month

get_stock_timeseries_data

Optional, default: "1h".  Supported intervals: 1min, 5min, 15min, 30min, 45min, 1h, 2h, 4h, 8h, 1day, 1week, 1month

Stock Time Series Search

Searches TwelveData for a stock symbol that matches the users request and retrieves stock symbol information like: symbol, instrument name, exchange, mic code, exchange timezone, instrument type, and country.

get_stock_symbol

Stock Symbol Search

Searches TwelveData for a stock symbol that matches the users request to retrieve the latest quote of the selected symbol.

get_stock_quote

Symbol ticker of the instrument.  (E.g. AAPL, DVN, EOG, ...)

Stock Symbol Quote

Retrieves pull request details from Azure Dev Ops given a project Id, respository Id, and pull request ID.

azure_dev_ops_get_pull_request

Azure DevOps - Get Pull Request Details

This access token will be provided by the user and allow the agent to interact with DevOps

DevOpsPersonalAccessToken

Generate documents and let user download as HTML, PDF, or DOCX.

generate_docx

CSS styles to apply to the rendered HTML.

The title of the document to be used as the file name.

The HTML content to convert to a DOCX and PDF file.

Document Generator

Create a PowerPoint presentation about the provided topic with image prompts and scripts.

create_powerpoint_slide

An array of slides containing titles, body texts, image prompts, and scripts.

Presentation Generator

Generate a customized AI image using Recraft's image generation capabilities. This function accepts a textual prompt, specifies the desired image style (either 'realistic_image' or 'digital_illustration'), defines the image size, and requires a substyle to refine the visual output. If none of the available substyles are suitable or the prompt is too vague, the 'substyle' should default to the value of 'style'. Suitable for creating high-quality images for presentations, marketing materials, or digital content creation.

generate_recraft_ai_image

A detailed textual description of the image to be generated. This guides the AI in creating an image that aligns with the user's vision.

The dimensions of the generated image in WxH format (e.g., '1280x1024').

A required substyle variant that enhances the main style. Ensure that the chosen substyle corresponds correctly to the selected 'style'.

**Behavior:** If none of the available substyles meet the user's request or the image prompt is too vague, set the value of 'substyle' to be the same as the selected 'style'.

**For `realistic_image`:** Choose one of the following based on the desired lighting and photo effect:
- `b_and_w` for black and white photos
- `enterprise` for a corporate/professional look
- `hard_flash` for harsh lighting
- `hdr` for high dynamic range
- `motion_blur` for action shots
- `natural_light` for outdoor/ambient lighting
- `studio_portrait` for professional portraits

**For `digital_illustration`:** Choose one of the following based on the desired artistic style:
- `2d_art_poster` / `2d_art_poster_2` for stylized promotional art and posters
- `engraving_color` for detailed etching-style with color
- `grain` for textured/noisy artistic effect
- `hand_drawn` for a natural sketched look
- `hand_drawn_outline` for emphasized line art
- `handmade_3d` for clay/sculpture-like 3D rendering
- `infantile_sketch` for children's drawing style
- `pixel_art` for 8-bit/16-bit retro gaming aesthetic

**Example Usage:** If `style` is set to `realistic_image` and the context is a landscape, select `natural_light` as the `substyle`. If no suitable substyle is available or the prompt is too vague, set `substyle` to `realistic_image`.

Recraft AI

Generate a customized AI image using Recraft's basic image generation capabilities. This function accepts a textual prompt, automatically determines the appropriate style ('realistic_image' or 'digital_illustration'), and defines the image size. Suitable for creating high-quality images for presentations, marketing materials, or digital content creation.

generate_recraft_ai_basic_image

Recraft AI (basic image)

query_snowflake_ehs

Snowflake Query EHS

Retrieves user metadata from Alation by user ID.

alation_user_by_id

Alation User By Id

Retrieves a document from Alation by document ID.

alation_document_by_id

Alation Document By Id

azure_dev_ops_get_wiki_page_detail

The Azure Dev Ops Project ID or Project Name

Azure DevOps - Get Wiki Page Details

Get Llama 3.3's perspective on the conversation or query

ask_llama

Array of conversation messages with roles and content

Ask Llama 3.3 70B

Get Gemini 2.0 Flash's perspective using Google's Vertex AI platform

ask_gemini

Array of conversation messages in Gemini format

Ask Gemini

Sends an email using SendGrid API with basic to, subject, and body parameters

send_email

Main content/body of the email. Can include plain text or HTML

SendGrid Email

Get GPT-4o's perspective on the conversation or query

ask_gpt4o

Ask GPT-4o

Get Claude 3 Sonnet's perspective on the conversation or query

ask_claude

Ask Claude

Perform a semantic search to get relevant content chunks of a specific document by Id from Devon's Cognitive Cache service. You must know the Document Id before calling this function. If needed perform a Cognitive Cache search to discover a specific document's Id.

search_cognitive_cache_document_by_id

Cognitive Cache Search Document By Id

Sends an email with Data@Devon agent feedback using SendGrid API with subject, and body parameters

Login to ChatDVN v3 Dev