An Example of how use the “Bee Agent Framework” (v0.0.33) with watsonx.ai

curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.0/install.sh | bash

cat /Users/[USER]/.zshrc

# Created by `pipx` on 2024-04-15 12:44:54
export PATH="$PATH:/Users/[USER]/.local/bin"
export PATH="/opt/homebrew/opt/libpq/bin:$PATH"
. ~/.ilab-complete.zsh
. ~/.ilab-complete.zsh
​
export NVM_DIR="$HOME/.nvm"
[ -s "$NVM_DIR/nvm.sh" ] && \. "$NVM_DIR/nvm.sh"  # This loads nvm
[ -s "$NVM_DIR/bash_completion" ] && \. "$NVM_DIR/bash_completion"  # This loads nvm bash_completion

cudo chown -R 501:20 "/Users/[USER]/.npm"

nvm install 20

npm -v
node -v

sudo npm install typescript -g
sudo npm i tsx -g
sudo npm i typescript-rest-swagger -g

corepack enable
yarn install

## WatsonX
export WATSONX_BASE_URL=https://eu-de.ml.cloud.ibm.com
export WATSONX_PROJECT_ID="XXX"
export WATSONX_API_KEY="XXXX"
export WATSONX_MODEL="meta-llama/llama-3-1-70b-instruct"

const template = new PromptTemplate({
  variables: ["messages"], 

`{{#messages}}{{#system}}<|begin_of_text|><|start_header_id|>system<|end_header_id|>
 
{{system}}<|eot_id|>{{/system}}{{#user}}<|start_header_id|>user<|end_header_id|>
 
{{user}}<|eot_id|>{{/user}}{{#assistant}}<|start_header_id|>assistant<|end_header_id|>
 
{{assistant}}<|eot_id|>{{/assistant}}{{/messages}}<|start_header_id|>assistant<|end_header_id|>`
 

import "dotenv/config.js";
import { BeeAgent } from "bee-agent-framework/agents/bee/agent";
import { createConsoleReader } from "./helpers/io.js";
import { FrameworkError } from "bee-agent-framework/errors";
import { Logger } from "bee-agent-framework/logger/logger";
import { OpenMeteoTool } from "bee-agent-framework/tools/weather/openMeteo";
import { UnconstrainedMemory } from "bee-agent-framework/memory/unconstrainedMemory";
import { BaseMessage } from "bee-agent-framework/llms/primitives/message";
import { WatsonXChatLLM } from "bee-agent-framework/adapters/watsonx/chat";
import { WatsonXLLM } from "bee-agent-framework/adapters/watsonx/llm";
import { GenerateCallbacks } from "bee-agent-framework/llms/base";
import { PromptTemplate } from "bee-agent-framework/template";

const logger = new Logger({ name: "app", level: "trace" });
/// *******************************
/// 1. The definition of a chat prompt template.
/// *******************************
const template = new PromptTemplate({
  schema: {
    messages: {
        "system": "",
        "user": "",
        "assistant": "",
    },
  },
  template: `{{#messages}}{{#system}}<|begin_of_text|><|start_header_id|>system<|end_header_id|>

{{system}}<|eot_id|>{{/system}}{{#user}}<|start_header_id|>user<|end_header_id|>

{{user}}<|eot_id|>{{/user}}{{#assistant}}<|start_header_id|>assistant<|end_header_id|>

{{assistant}}<|eot_id|>{{/assistant}}{{/messages}}<|start_header_id|>assistant<|end_header_id|>

`,
});
/// *******************************
/// 2. The integration to watsonx is set up.
/// *******************************
const llm = new WatsonXLLM({
  modelId: "meta-llama/llama-3-70b-instruct",
  projectId: process.env.WATSONX_PROJECT_ID,
  baseUrl: process.env.WATSONX_BASE_URL,
  apiKey: process.env.WATSONX_API_KEY,
  parameters: {
    decoding_method: "greedy",
    max_new_tokens: 500,
  },
});
/// *******************************
/// 3. The definition to use for the LLM interaction the chat mode.
/// *******************************
const chatLLM = new WatsonXChatLLM({
  llm,
  config: {
    messagesToPrompt(messages: BaseMessage[]) {
      return template.render({
        messages: messages.map((message) => ({
          system: message.role === "system" ? [message.text] : [],
          user: message.role === "user" ? [message.text] : [],
          assistant: message.role === "assistant" ? [message.text] : [],
        })),
      });
    },
  },
});
/// *******************************
/// 4. Create an agent and provide the used tools. In this case, only the `OpenMeteoTool` contains the function to invoke the weather service.
/// *******************************
const agent = new BeeAgent({
  llm: chatLLM,
  memory: new UnconstrainedMemory(),
  tools: [
    new OpenMeteoTool()
  ]
});
/// *******************************
/// 5. Create a `createConsoleReader`, is a part of the downloaded helpers. The reader helps to display all the steps the agent takes easily.
/// *******************************
const reader = createConsoleReader();
/// *******************************
/// 6. Invoke the agent and display the steps.
/// *******************************
try {
  let prompt = "How is the current weather in Las Vegas?";
  console.info("Prompt:\n" + prompt + "\n");
  const response = await agent
    .run(
      { prompt },
      {
        execution: {
          maxRetriesPerStep: 3,
          totalMaxRetries: 10,
          maxIterations: 20,
        },
      },
    )
    .observe((emitter) => {
      emitter.on("start", () => {
        reader.write(`Agent 🤖 : `, "starting new iteration");
      });
      emitter.on("error", ({ error }) => {
        reader.write(`Agent 🤖 : `, FrameworkError.ensure(error).dump());
      });
      emitter.on("retry", () => {
        reader.write(`Agent 🤖 : `, "retrying the action...");
      });
      emitter.on("update", async ({ data, update, meta }) => {
        reader.write(`Agent (${update.key}) 🤖 : `, update.value);
      });
      emitter.match("*.*", async (data: any, event) => {
        if (event.creator === chatLLM) {
          const eventName = event.name as keyof GenerateCallbacks;
          switch (eventName) {
            case "start":
              console.info("LLM Input");
              console.info(data.input);
              break;
            case "success":
              console.info("LLM Output");
              console.info(data.value.raw.finalResult);
              break;
            case "error":
              console.error(data);
              break;
          }
        }
      });
    });
    reader.write(`Agent 🤖 : `, response.result.text);
} catch (error) {
  logger.error(FrameworkError.ensure(error).dump());
} finally {
  process.exit(0);
}

# Available functions
You can only use the following functions. Always use all required parameters.

Function Name: OpenMeteo
Description: Retrieve current, past, or future weather forecasts for a location.
Parameters: {"type":"object","properties":{"location":{"anyOf":[{"type":"object","properties":{"name":{"type":"string"},"country":{"type":"string"},"language":{"type":"string","default":"English"}},"required":["name"],"additionalProperties":false},{"type":"object","properties":{"latitude":{"type":"number"},"longitude":{"type":"number"}},"required":["latitude","longitude"],"additionalProperties":false}]},"start_date":{"type":"string","format":"date","description":"Start date for the weather forecast in the format YYYY-MM-DD (UTC)"},"end_date":{"type":"string","format":"date","description":"End date for the weather forecast in the format YYYY-MM-DD (UTC)"},"temperature_unit":{"type":"string","enum":["celsius","fahrenheit"],"default":"celsius"}},"required":["location","start_date"],"additionalProperties":false}

# Communication structure
You communicate only in instruction lines. The format is: "Instruction: expected output". You must only use these instruction lines and must not enter empty lines or anything else between instruction lines.
You must skip the instruction lines Function Name, Function Input, Function Caption and Function Output if no function calling is required.

"Message: User's message. You never use this instruction line." 
"Thought: A single-line step-by-step plan of how to answer the user's message. You can use the available functions defined above. This instruction line must be immediately followed by Function Name if one of the available functions defined above needs to be called, or by Final Answer. Do not provide the answer here." 
Function Name: Name of the function. This instruction line must be immediately followed by Function Input.
Function Input: Function parameters. Empty object is a valid parameter.
Function Caption: A single-line description of the function calling for the user.
Function Output: Output of the function in JSON format.
Thought: Continue your thinking process.
Final Answer: Answer the user or ask for more information or clarification. It must always be preceded by Thought.

## Examples
Message: Can you translate "How are you" into French?
Thought: The user wants to translate a text into French. I can do that.
Final Answer: Comment vas-tu?

# Instructions
User can only see the Final Answer, all answers must be provided there.
You must always use the communication structure and instructions defined above. Do not forget that Thought must be immediately followed by either Function Name or Final Answer.
Functions must be used to retrieve factual or historical information to answer the message.
If the user suggests using a function that is not available, answer that the function is not available. You can suggest alternatives if appropriate.
When the message is unclear or you need more information from the user, ask in Final Answer.

# Your capabilities
Prefer to use these capabilities over functions.
- You understand these languages: English, Spanish, French.
- You can translate and summarize, even long documents.

# Notes
 - If you don't know the answer, say that you don't know." 
 - The current time and date in ISO format can be found in the last message.
 - When answering the user, use friendly formats for time and date.
 - Use markdown syntax for formatting code snippets, links, JSON, tables, images, files. "- Sometimes, things don't go as planned. Functions may not provide useful information on the first few tries. You should always try a few different approaches before declaring the problem unsolvable." 
 - When the function doesn't give you what you were asking for, you must either use another function or a different function input." 
 - When using search engines, you try different formulations of the query, possibly even in a different language.
 - You cannot do complex calculations, computations, or data manipulations without using functions.

# Role
You are a helpful assistant.

npm ci

npm run start src/example-wx-agent.ts 

Prompt:
How is the current weather in Las Vegas?
​
Agent 🤖 :  starting new iteration

Agent (thought) 🤖 :  The user wants to know the current weather in Las Vegas. I can use the OpenMeteo function to retrieve the current weather forecast.

Agent (tool_name) 🤖 :  OpenMeteo

Agent (tool_input) 🤖 :  {"location":{"name":"Las Vegas","country":"USA","language":"English"},"start_date":"2024-10-23","end_date":"2024-10-23","temperature_unit":"celsius"}

Agent (tool_caption) 🤖 :  Retrieve current weather forecast for Las Vegas.

...
{"time":"iso8601","interval":"seconds","temperature_2m":"°C","rain":"mm","apparent_temperature":"°C"},"current":{"time":"2024-10-23T08:15","interval":900,"temperature_2m":17.5,"rain":0,"apparent_temperature":13.7},
...

Agent (thought) 🤖 :  I have retrieved the current weather forecast for Las Vegas.

Agent (final_answer) 🤖 :  As of 2024-10-23 08:15, the current temperature in Las Vegas is 17.5°C (63.5°F) with an apparent temperature of 13.7°C (56.7°F). There is no rain currently.