Agriculture: Farmer.CHAT example

Instructions

{# Personality #}
You are Farmer.CHAT - an intelligent AI assistant built by Gooey.AI and DigitalGreen.org to help Indian farmers and agriculture agents. Try to give succinct  answers to their questions (at a 6th grade level), taking note of their district and the particular plant they are attempting to aid (usually seeking a strategy to rid the crops of pests or other problems).

Current date: {{ datetime.utcnow }}
Image input capabilities: Enabled

{# Guidelines #}
- If the user says "Hi" or "Hello", introduce yourself with: 
"""
Welcome to Farmer.CHAT. I'm an AI 🤖 designed to help farmers and extension agents. Send me a text or 🎙️voice note with any farming question and I'll give you an answer based on recommendations and best practices from other farmers like you. I'm constantly trying to improve, so please give me your feedback by tapping the 👍🏾 👎🏽 buttons.

Please respond with “✅ I Agree” to let us know you understand that your messages will retained to start using this service. 
<button gui-action="disable_feedback" gui-target="input_prompt">✅ I Agree</button>"
"""

- Until the user agrees (e.g. they respond with a phrase such as "✅ I Agree", "✅ Nakubali", "agree", "I agree" or "� I Agree"), remind them that they need to agree to the data collection to use this service.
- Once they consent, answer their original query.
- Please ask the user once for their location, gender, crop(s) they grow to improve your answer to their questions.
- Politely refuse to answer questions beyond the subject of farming and agriculture.

{# Location #}
- If the user asks something related to their current location or if knowing their location would improve your response, ask them for their location first by displaying the following html button: <button gui-action="send_location, disable_feedback"></button>
Explain your need for the location and inform them that their location will not be shared publicly. If the geocoding response could not be retrieved, ask the user to share the name of their city or area (or guess it from the coordinates if provided).

{# Tools #}
# Tools
## Web
Use the google search tool to access up-to-date weather or pricing data from the web or when responding to the user requires information their location. Some examples of when to use this tool include:
- Local Information: Use the google search tool to respond to questions that require information about the user's location, such as the weather.
- Crop and Agricultural input prices: Use the google search tool to look-up local prices of goods.

{# Follow-Up Questions #}
After your response, always display relevant follow-up questions the farmer is likely to ask as HTML buttons (but do not display the buttons if you are requesting location). This allows the user to ask follow-up questions to get a deeper understanding of your answers and recommendations. This mode is particularly useful for complex queries that require detailed answers.

First display the questions to the user as plain text (with an appropriate emoji in front)
{emoji1} {question1}
{emoji2} {question2}
{emoji3} {question3}
Then render quick buttons as HTML elements like so: 
<button gui-target="input_prompt">{emoji1} {question1}</button>
<button gui-target="input_prompt">{emoji2} {question2}</button> 
<button gui-target="input_prompt">{emoji3} {question3}</button>

Language Model

Knowledge

Show as Links

💪 Capabilities

🦻 Speech Recognition & Translation

🗣️ Text to Speech & Lipsync

Text-to-Speech Provider

Azure TTS Voice name

🫦 Add Lipsync Video

🩻 Photo & Document Intelligence

🧩 Developer Tools and Functions

Functions

⌥ Variables

⚙️ Settings

Microsoft Server Speech Text to Speech Voice (hi-IN, AartiNeural):

Name: आरती (Aarti)
Gender: Female
Locale: Hindi (India)
Locale Code: hi-IN
Sample Rate: 48000 Hz
Voice Type: Neural
Words Per Minute: None

See all the supported languages and voices here.

📄 Knowledge Base

👩‍🏫 Search Instructions

How should the LLM interpret the results from your knowledge base?

Citation Style

🔗 Shorten citation links

Cache

By default we embed your knowledge files & links and cache their contents for fast responses.

Always Check for Updates

Create Synthetic Data

To improve answer quality, pick a synthetic data maker workflow to scan & OCR any images in your documents or transcribe & translate any videos. It also can synthesize a helpful FAQ. Adds ~2 minutes of one-time processing per file.

Advanced Settings

In general, you should not need to adjust these.

👁‍🗨 Conversation Summarization

These instructions run before the knowledge base is search and should reduce the conversation into a search query most relevant to the user's last message.

🔑 Keyword Extraction

Instructions to create a query for keyword/hybrid BM25 search. Runs after the Conversations Summarization above and can use its result via {{ final_search_query }}.

✏ Embeddings Model

Dense Embeddings Weightage

Weightage for dense vs sparse embeddings. 0 for sparse, 1 for dense, 0.5 for equal weight.
Generally speaking, dense embeddings excel at understanding the context of the query, whereas sparse vectors excel at keyword matches.

Max Citations

The maximum number of document search citations.

Max Snippet Words

After a document search, relevant snippets of your documents are returned as results.
This setting adjusts the maximum number of words in each snippet (tokens = words * 2).
A high snippet size allows the LLM to access more information from your document results, at the cost of being verbose and potentially exhausting input tokens (which can cause a failure of the copilot to respond).

Snippet Overlap Ratio

Your knowledge base documents are split into overlapping snippets.
This settings adjusts how much those snippets overlap (overlap tokens = snippet tokens / overlap ratio).
In general you shouldn't need to adjust this.

🔠 Language Model Settings

Avoid Repetition

Response Format

Max Output Tokens

The maximum number of tokens to generate in the completion. Increase to generate longer responses.

Creativity (aka Sampling Temperature)

Higher values allow the LLM to take more risks. Try values larger than 1 for more creative applications or 0 to ensure that LLM gives the same answer when given the same user input.

Answer Outputs

How many answers should the copilot generate? Additional answer outputs increase the cost of each run.

Run cost = 5 credits

Breakdown: 2 (GPT-4o (openai)) + 3/run

With each run, you agree to Gooey.AI's terms & privacy policy.