### Register Tools with an Agent

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/tools.html

Demonstrates how to register multiple tools with an agent using the `register_tools` method. This example registers weather-related tools for a weather assistant agent.

```r
library(mini007)

openai_4_1 <- ellmer::chat(
  name = "openai/gpt-4.1",
  credentials = function() {Sys.getenv("OPENAI_API_KEY")},
  echo = "none"
)

weather_agent <- Agent$new(
  name = "weather_assistant",
  instruction = "You are a weather assistant.",
  llm_object = openai_4_1
)

weather_function_algiers <- function() {
  msg <- glue::glue(
    "35 degrees Celcius, it's sunny and there's no precipitation."
  )
  msg
}

get_weather_in_algiers <- ellmer::tool(
  fun = weather_function_algiers,
  name = "get_weather_in_algiers",
  description = "Provide the current weather in Algiers, Algeria."
)

weather_function_berlin <- function() {
  msg <- glue::glue(
    "10 degrees Celcius, it's cold"
  )
  msg
}

get_weather_in_berlin <- ellmer::tool(
  fun = weather_function_berlin,
  name = "get_weather_in_berlin",
  description = "Provide the current weather in Berlin, Germany"
)

weather_agent$register_tools(
  tools = list(
    get_weather_in_algiers, 
    get_weather_in_berlin
  )
)
```

--------------------------------

### Install mini007 Package

Source: https://github.com/feddelegrand7/mini007/blob/main/README.md

Install the mini007 package from CRAN. Ensure you have R installed.

```r
install.packages("mini007")
```

--------------------------------

### Initialize ClipboardJS

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/leadagent.html

Initializes ClipboardJS for copying code. It handles both standard copy buttons and those within Quarto modals, using the 'getTextToCopy' function to get the content.

```javascript
const clipboard = new window.ClipboardJS('.code-copy-button:not([data-in-quarto-modal])', {
 text: getTextToCopy
});
clipboard.on('success', onCopySuccess);
if (window.document.getElementById('quarto-embedded-source-code-modal')) {
 const clipboardModal = new window.ClipboardJS('.code-copy-button[data-in-quarto-modal]', {
 text: getTextToCopy,
 container: window.document.getElementById('quarto-embedded-source-code-modal')
 });
clipboardModal.on('success', onCopySuccess);
}
```

--------------------------------

### Manage Agents with LeadAgent

Source: https://context7.com/feddelegrand7/mini007/llms.txt

Demonstrates how to initialize agents, register them to a LeadAgent, and manage the agent collection through registration, removal, and clearing.

```r
openai_4_1_mini <- ellmer::chat(
  name = "openai/gpt-4.1-mini",
  api_key = Sys.getenv("OPENAI_API_KEY"),
  echo = "none"
)

researcher <- Agent$new(
  name = "researcher",
  instruction = "You are a research assistant.",
  llm_object = openai_4_1_mini
)

summarizer <- Agent$new(
  name = "summarizer",
  instruction = "You summarize text.",
  llm_object = openai_4_1_mini
)

translator <- Agent$new(
  name = "translator",
  instruction = "You translate to German.",
  llm_object = openai_4_1_mini
)

lead_agent <- LeadAgent$new(name = "Leader", llm_object = openai_4_1_mini)

# Register agents
lead_agent$register_agents(c(researcher, summarizer, translator))
# Agent(s) successfully registered.

# View registered agents
lapply(lead_agent$agents, function(x) x$name)
# [[1]] [1] "researcher"
# [[2]] [1] "summarizer"
# [[3]] [1] "translator"

# Remove specific agent by ID
lead_agent$remove_agents(translator$agent_id)

lapply(lead_agent$agents, function(x) x$name)
# [[1]] [1] "researcher"
# [[2]] [1] "summarizer"

# Clear all agents
lead_agent$clear_agents()

length(lead_agent$agents)
# [1] 0
```

--------------------------------

### Highlight Code Lines

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/leadagent.html

Selects and highlights specific lines of code based on annotation data. This is used for interactive code examples.

```javascript
let selectedAnnoteEl;
const selectorForAnnotation = ( cell, annotation) => {
    let cellAttr = 'data-code-cell="' + cell + '"';
    let lineAttr = 'data-code-annotation="' + annotation + '"';
    const selector = 'span[' + cellAttr + '][' + lineAttr + ']';
    return selector;
}

const selectCodeLines = (annoteEl) => {
    const doc = window.document;
    const targetCell = annoteEl.getAttribute("data-target-cell");
    const targetAnnotation = annoteEl.getAttribute("data-target-annotation");
    const annoteSpan = window.document.querySelector(selectorForAnnotation(targetCell, targetAnnotation));
    const lines = annoteSpan.getAttribute("data-code-lines").split(",");
    const lineIds = lines.map((line) => {
        return targetCell + "-" + line;
    })
    let top = null;
    let height = null;
    let parent = null;
    if (lineIds.length > 0) {
        //compute the position of the single el (top and bottom and make a div)
        const el = window.document.getElementById(lineIds[0]);
        top = el.offsetTop;
        height = el.offsetHeight;
        parent = el.parentElement.parentElement;
        if (lineIds.length > 1) {
            const lastEl = window.document.getElementById(lineIds[lineIds.length - 1]);
            const bottom = lastEl.offsetTop + lastEl.offsetHeight;
            height = bottom - top;
        }
        if (top !== null && height !== null && parent !== null) {
            // cook up a div (if necessary) and position it
            let div = window.document.getElementById("code-annotation-line-highlight");
            if (div === null) {
                div = window.document.createElement("div");
                div.setAttribute("id", "code-annotat
```

--------------------------------

### Visualize Orchestration Plan with LeadAgent$visualize_plan()

Source: https://context7.com/feddelegrand7/mini007/llms.txt

Generates a diagram of the agent sequence. Requires the DiagrammeR package to render the output.

```r
openai_4_1_mini <- ellmer::chat(
  name = "openai/gpt-4.1-mini",
  api_key = Sys.getenv("OPENAI_API_KEY"),
  echo = "none"
)

researcher <- Agent$new(
  name = "researcher",
  instruction = "You are a research assistant.",
  llm_object = openai_4_1_mini
)

summarizer <- Agent$new(
  name = "summarizer",
  instruction = "You summarize text.",
  llm_object = openai_4_1_mini
)

lead_agent <- LeadAgent$new(name = "Leader", llm_object = openai_4_1_mini)
lead_agent$register_agents(c(researcher, summarizer))

# Generate a plan first
lead_agent$generate_plan("Research quantum computing and summarize it")

# Visualize the plan (requires DiagrammeR package)
lead_agent$visualize_plan()
# Displays a left-to-right diagram: [researcher] -> [summarizer]
# Hovering shows the delegated prompt for each agent
```

--------------------------------

### Agent Conversation Reset

Source: https://context7.com/feddelegrand7/mini007/llms.txt

Clear the conversation history while preserving the current system prompt. Starts a fresh conversation without recreating the agent.

```APIDOC
## Agent$reset_conversation_history()

### Description
Clear the conversation history while preserving the current system prompt. Starts a fresh conversation without recreating the agent.

### Method
`reset_conversation_history()`

### Parameters
None

### Request Example
```r
# Reset the conversation
agent$reset_conversation_history()
```

### Response Example
- Confirmation message indicating that the conversation history has been reset and the system prompt is preserved.
```

--------------------------------

### Code Annotation Helper Function

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/leadagent.html

A helper function to determine if an element is part of a code annotation. It checks if any of the element's classes start with 'code-annotation-'.

```javascript
const isCodeAnnotation = (el) => {
 for (const clz of el.classList) {
 if (clz.startsWith('code-annotation-')) {
 return true;
 }
 }
 return false;
}
```

--------------------------------

### Create an Agent instance

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/agents.html

Instantiate an Agent object by providing a name, system instructions, and the underlying LLM object.

```R
polar_bear_researcher <- Agent$new(
  name = "POLAR BEAR RESEARCHER",
  instruction = "You are an expert in polar bears, you task is to collect information about polar bears. Answer in 1 sentence max.",
  llm_object = openai_4_1_mini
)
```

--------------------------------

### Initialize OpenAI Chat Client

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/leadagent.html

Sets up an ellmer chat object using an OpenAI API key retrieved from environment variables.

```R
library(mini007)

retrieve_open_ai_credential <- function() {
  Sys.getenv("OPENAI_API_KEY")
}

openai_4_1_mini <- ellmer::chat(
  name = "openai/gpt-4.1-mini",
  credentials = retrieve_open_ai_credential,
  echo = "none"
)
```

--------------------------------

### Initialize and Register Agents with LeadAgent

Source: https://context7.com/feddelegrand7/mini007/llms.txt

Create specialized agents and register them with a LeadAgent to enable automated task decomposition.

```r
library(mini007)

openai_4_1_mini <- ellmer::chat(
  name = "openai/gpt-4.1-mini",
  api_key = Sys.getenv("OPENAI_API_KEY"),
  echo = "none"
)

# Create specialized agents
researcher <- Agent$new(
  name = "researcher",
  instruction = "You are a research assistant. Answer factual questions with detailed and accurate information. Do not answer with more than 2 lines.",
  llm_object = openai_4_1_mini
)

summarizer <- Agent$new(
  name = "summarizer",
  instruction = "You are an agent designed to summarise a given text into 3 distinct bullet points.",
  llm_object = openai_4_1_mini
)

translator <- Agent$new(
  name = "translator",
  instruction = "Your role is to translate a text from English to German",
  llm_object = openai_4_1_mini
)

# Create the LeadAgent (no instruction needed - it has built-in task decomposition)
lead_agent <- LeadAgent$new(
  name = "Leader",
  llm_object = openai_4_1_mini
)

# Register the specialized agents
lead_agent$register_agents(c(researcher, summarizer, translator))
# Agent(s) successfully registered.

# View registered agents
lapply(lead_agent$agents, function(x) x$name)
# [[1]] [1] "researcher"
# [[2]] [1] "summarizer"
# [[3]] [1] "translator"
```

--------------------------------

### Get Color Scheme Sentinel

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/index.html

Retrieves the current color scheme preference from localStorage or a local variable. Returns 'alternate' for dark mode, 'default' for light mode, or null if not set.

```javascript
const getColorSchemeSentinel = () => { if (!isFileUrl()) { const storageValue = window.localStorage.getItem("quarto-color-scheme"); return storageValue != null ? storageValue : localAlternateSentinel; } else { return localAlternateSentinel; } }
```

--------------------------------

### Create and Invoke an Agent

Source: https://context7.com/feddelegrand7/mini007/llms.txt

Demonstrates creating an Agent with a specific role and LLM, then invoking it with a prompt. Accesses agent ID and conversation history.

```r
library(mini007)

# Create an LLM object using ellmer
openai_4_1_mini <- ellmer::chat(
  name = "openai/gpt-4.1-mini",
  api_key = Sys.getenv("OPENAI_API_KEY"),
  echo = "none"
)

# Create an Agent with a specific role
polar_bear_researcher <- Agent$new(
  name = "POLAR BEAR RESEARCHER",
  instruction = "You are an expert in polar bears. Answer in 1 sentence max.",
  llm_object = openai_4_1_mini
)

# Each agent has a unique ID
polar_bear_researcher$agent_id
# [1] "a1b2c3d4-e5f6-7890-abcd-ef1234567890"

# Invoke the agent with a prompt
response <- polar_bear_researcher$invoke("Are polar bears dangerous for humans?")
# [1] "Yes, polar bears can be extremely dangerous to humans..."

# Access conversation history
polar_bear_researcher$messages
# [[1]]
# $role
# [1] "system"
# $content
# [1] "You are an expert in polar bears..."
# [[2]]
# $role
# [1] "user"
# $content
# [1] "Are polar bears dangerous for humans?"
# [[3]]
# $role
# [1] "assistant"
# $content
# [1] "Yes, polar bears can be extremely dangerous..."
```

--------------------------------

### Configure and Register LeadAgent

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/leadagent.html

Initializes the LeadAgent and registers the previously created agents to it for task orchestration.

```R
lead_agent <- LeadAgent$new(
  name = "Leader", 
  llm_object = openai_4_1_mini
)

lead_agent$register_agents(c(researcher, summarizer, translator))

lapply(lead_agent$agents, function(x) {x$name})
```

--------------------------------

### Get Text to Copy for Clipboard

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/leadagent.html

Retrieves the inner text of a code element, excluding any code annotation elements, for clipboard copying. It clones the parent element to avoid modifying the original DOM.

```javascript
const getTextToCopy = function(trigger) {
 const outerScaffold = trigger.parentElement.cloneNode(true);
 const codeEl = outerScaffold.querySelector('code');
 for (const childEl of codeEl.children) {
 if (isCodeAnnotation(childEl)) {
 childEl.remove();
 }
 }
 return codeEl.innerText;
}
```

--------------------------------

### Agent Class Initialization

Source: https://context7.com/feddelegrand7/mini007/llms.txt

Defines how to instantiate a new Agent with a specific role, instruction, and LLM object.

```APIDOC
## Agent$new()

### Description
Creates a new instance of an Agent with a unique identity and persistent message history.

### Parameters
- **name** (string) - Required - The name of the agent.
- **instruction** (string) - Required - The system prompt or role definition for the agent.
- **llm_object** (object) - Required - An ellmer chat model object.

### Request Example
agent <- Agent$new(name = "Researcher", instruction = "You are an expert.", llm_object = openai_model)
```

--------------------------------

### CEO2's Second Response

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/leadagent.html

The second CEO agent rejects CEO1's high-budget proposal due to cost constraints and suggests a phased, cost-effective strategy starting with digital marketing and micro-influencers.

```text
Your large-scale, high-budget approach clearly conflicts with my core 
constraint of maintaining a low marketing spend; therefore, I counterpropose a 
phased strategy that begins with focused digital marketing and micro-influencer
collaborations to build brand awareness cost-effectively in Germany, with 
potential gradual scaling depending on export growth and budget availability 
over the next two years.
```

--------------------------------

### Get Color Scheme Sentinel

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/NEWS.html

Retrieves the current color scheme preference from local storage or a local variable. Returns 'alternate' if the alternate theme is set, otherwise 'default'. Handles file URLs differently.

```javascript
const getColorSchemeSentinel = () => {
if (!isFileUrl()) {
const storageValue = window.localStorage.getItem("quarto-color-scheme");
return storageValue != null ? storageValue : localAlternateSentinel;
} else {
return localAlternateSentinel;
}
}
```

--------------------------------

### Initialize CEO Agents and Lead Agent

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/leadagent.html

Initializes two CEO agents with distinct instructions regarding marketing strategy and budget, and a LeadAgent to manage their dialogue. Uses the openai_4_1_mini LLM object.

```R
ceo1 <- Agent$new(
  name = "ceo1",
  instruction = paste0(
    "You are a CEO in  a dates company based in Ouergla, Algeria, ", 
    "You want to boost their exports to Germany. ", 
    "You don't care about the budget. You want to spend as much as possible. "
  ),
  llm_object = openai_4_1_mini
)

ceo2 <- Agent$new(
  name = "ceo2",
  instruction = paste0(
    "You are the CEO of a dates company based in Ouergla, Algeria. ", 
    "You are considering starting a marketing compaign to boost the exports to Germany. ",
    "For you the marketing budget is super important and you don't want to spend too much. "
  ),
  llm_object = openai_4_1_mini
)

lead_agent <- LeadAgent$new(
  name = "Leader",
  llm_object = openai_4_1_mini
)
```

--------------------------------

### Initialize Quarto Document UI

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/testing.html

Configures document-level UI elements including color scheme toggles, anchor links, and clipboard copy functionality.

```JavaScript
window.document.addEventListener("DOMContentLoaded", function (event) { // Ensure there is a toggle, if there isn't float one in the top right if (window.document.querySelector('.quarto-color-scheme-toggle') === null) { const a = window.document.createElement('a'); a.classList.add('top-right'); a.classList.add('quarto-color-scheme-toggle'); a.href = ""; a.onclick = function() { try { window.quartoToggleColorScheme(); } catch {} return false; }; const i = window.document.createElement("i"); i.classList.add('bi'); a.appendChild(i); window.document.body.appendChild(a); } setColorSchemeToggle(hasAlternateSentinel()) const icon = ""; const anchorJS = new window.AnchorJS(); anchorJS.options = { placement: 'right', icon: icon }; anchorJS.add('.anchored'); const isCodeAnnotation = (el) => { for (const clz of el.classList) { if (clz.startsWith('code-annotation-')) { return true; } } return false; } const onCopySuccess = function(e) { // button target const button = e.trigger; // don't keep focus button.blur(); // flash "checked" button.classList.add('code-copy-button-checked'); var currentTitle = button.getAttribute("title"); button.setAttribute("title", "Copied!"); let tooltip; if (window.bootstrap) { button.setAttribute("data-bs-toggle", "tooltip"); button.setAttribute("data-bs-placement", "left"); button.setAttribute("data-bs-title", "Copied!"); tooltip = new bootstrap.Tooltip(button, { trigger: "manual", customClass: "code-copy-button-tooltip", offset: [0, -8]}); tooltip.show(); } setTimeout(function() { if (tooltip) { tooltip.hide(); button.removeAttribute("data-bs-title"); button.removeAttribute("data-bs-toggle"); button.removeAttribute("data-bs-placement"); } button.setAttribute("title", currentTitle); button.classList.remove('code-copy-button-checked'); }, 1000); // clear code selection e.clearSelection(); } const getTextToCopy = function(trigger) { const outerScaffold = trigger.parentElement.cloneNode(true); const codeEl = outerScaffold.querySelector('code'); for (const childEl of codeEl.children) { if (isCodeAnnotation(childEl)) { childEl.remove(); } } return codeEl.innerText; } const clipboard = new window.ClipboardJS('.code-copy-button:not([data-in-quarto-modal])', { text: getTextToCopy }); clipboard.on('success', onCopySuccess); if (window.document.getElementById('quarto-embedded-source-code-modal')) { const clipboardModal = new window.ClipboardJS('.code-copy-button[data-in-quarto-modal]', { text: getTextToCopy, container: window.document.getElementById('quarto-embedded-source-code-modal') }); clipboardModal.on('success', onCopySuccess); } var localhostRegex = new RegExp(/^(?:http|https):\/\/localhost\:?[0-9]*\//); var mailtoRegex = new RegExp(/^mailto:/); var filterRegex = new RegExp('/' + window.location.host + '/'); var isInternal = (href) => { return filterRegex.test(href) || localhostRegex.test(href) || mailtoRegex.test(href); } // Inspect non-navigation links and adorn them if external var links = window.document.querySelectorAll('a[href]:not(.nav-link):not(.navbar-brand):not(.toc-action):not(.sidebar-link):not(.sidebar-item-toggle):not(.pagination-link):not(.no-external):not([aria-hidden]):not(.dropdown-item):not(.quarto-navigation-tool):not(.about-link)'); for (var i=0; i<links.length; i++) { const link = links[i]; if (!isInternal(link.href)) { // undo the damage that might have been done by quarto-nav.js in the case of // links that we want to consider external if (link.dataset.originalHref !== undefined) { link.href = link.dataset.originalHref; } } } function tippyHover(el, contentFn, onTriggerFn, onUntriggerFn) { const config = { allowHTML: true, maxWidth: 500, delay: 100, arrow: false, appendTo: function(el) { return el.parentElement; }, interactive: true, interactiveBorder: 10, theme: 'quarto', placement: 'bottom-start', }; if (contentFn) { config.content = contentFn; } if (onTriggerFn) { config.onTrigger = onTriggerFn; } if (onUntriggerFn) { config.onUntrigger = onUntriggerFn; } window.tippy(el, config); } const noterefs = window.document.querySelectorAll('a[role="doc-noteref"]'); for (var i=0; i<noterefs.length; i++) { const ref = noterefs[i]; tippyHover(ref, function() { // use id or data attribute instead here let href = ref.getAttribute('data-footnote-href') || ref.getAttribute('href'); try { href = new URL(href).hash; } catch {} const id = href.replace(/^#\/?/, ""); const note = window.document.getElementById(id); if (note) { return note.innerHTML; } els
```

--------------------------------

### Initialize an LLM object for an Agent

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/agents.html

Configure an OpenAI chat object using the ellmer package to serve as the foundation for an agent.

```R
# no need to provide the system prompt, it will be set when creating the
# agent (see the 'instruction' parameter)

retrieve_open_ai_credential <- function() {
  Sys.getenv("OPENAI_API_KEY")
}

openai_4_1_mini <- ellmer::chat(
  name = "openai/gpt-4.1-mini",
  credentials = retrieve_open_ai_credential, 
  echo = "none"
)
```

--------------------------------

### LeadAgent$visualize_plan()

Source: https://context7.com/feddelegrand7/mini007/llms.txt

Visualizes the current orchestration plan as a diagram showing the sequence of agents.

```APIDOC
## GET LeadAgent$visualize_plan()

### Description
Visualizes the orchestration plan as a diagram showing the sequence of agents. Each agent node displays the assigned prompt on hover.
```

--------------------------------

### LeadAgent Class Initialization and Agent Registration

Source: https://context7.com/feddelegrand7/mini007/llms.txt

Initialize the LeadAgent and register specialized agents to it. The LeadAgent coordinates these agents for complex task decomposition.

```APIDOC
## LeadAgent Class

### Description
The `LeadAgent` class extends `Agent` to coordinate a group of specialized agents. It decomposes complex prompts into subtasks using LLMs and assigns each subtask to the most suitable registered agent, handling response chaining where each agent can consider prior results.

### Method
`LeadAgent$new(name, llm_object)` and `register_agents(agents)`

### Parameters
#### `LeadAgent$new`
- **name** (string) - Required - The name of the LeadAgent.
- **llm_object** (object) - Required - The language model object to be used by the LeadAgent.

#### `register_agents`
- **agents** (list of Agent objects) - Required - A list of specialized Agent objects to be registered with the LeadAgent.

### Request Example
```r
# Create specialized agents
researcher <- Agent$new(...)
summarizer <- Agent$new(...)
translator <- Agent$new(...)

# Create the LeadAgent
lead_agent <- LeadAgent$new(name = "Leader", llm_object = openai_4_1_mini)

# Register the specialized agents
lead_agent$register_agents(c(researcher, summarizer, translator))
```

### Response Example
- **register_agents**: Confirmation message indicating successful agent registration.
```

--------------------------------

### Generate Execution Plan with LeadAgent

Source: https://context7.com/feddelegrand7/mini007/llms.txt

Use generate_plan to preview how a prompt will be decomposed into subtasks across registered agents without executing them.

```r
openai_4_1_mini <- ellmer::chat(
  name = "openai/gpt-4.1-mini",
  api_key = Sys.getenv("OPENAI_API_KEY"),
  echo = "none"
)

researcher <- Agent$new(
  name = "researcher",
  instruction = "You are a research assistant. Answer factual questions.",
  llm_object = openai_4_1_mini
)

summarizer <- Agent$new(
  name = "summarizer",
  instruction = "You summarize text into bullet points.",
  llm_object = openai_4_1_mini
)

translator <- Agent$new(
  name = "translator",
  instruction = "You translate English to German.",
  llm_object = openai_4_1_mini
)

lead_agent <- LeadAgent$new(name = "Leader", llm_object = openai_4_1_mini)
lead_agent$register_agents(c(researcher, summarizer, translator))

# Generate a plan before execution
prompt <- "Tell me about the economic situation in Algeria, summarize it in 3 bullet points, then translate it into German."

plan <- lead_agent$generate_plan(prompt)
# Plan successfully generated.

# View the plan
plan
# [[1]]
# $agent_id [1] "abc123..."
# $agent_name [1] "researcher"
# $model_provider [1] "OpenAI"
# $model_name [1] "gpt-4.1-mini"
# $prompt [1] "Tell me about the economic situation in Algeria"
# [[2]]
# $agent_id [1] "def456..."
# $agent_name [1] "summarizer"
# $prompt [1] "Summarize the information in 3 bullet points"
# [[3]]
```

--------------------------------

### Initialize a Testing Agent

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/testing.html

Create an Agent instance configured with an LLM to act as a judge for response validation.

```R
library(mini007)

retrieve_open_ai_credential <- function() {
  Sys.getenv("OPENAI_API_KEY")
}

openai_4_1_mini <- ellmer::chat(
  name = "openai/gpt-4.1-mini",
  credentials = retrieve_open_ai_credential,
  echo = "none"
)

tester <- Agent$new(
  name = "tester",
  instruction = "You are agent responsible of testing LLM responses",
  llm_object = openai_4_1_mini
)
```

--------------------------------

### Initialize Tippy.js Tooltips for Cross-References

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/CODE_OF_CONDUCT.html

Initializes tippy.js tooltips for all elements with the class 'quarto-xref'. It handles fetching cross-reference content dynamically and displaying it in the tooltip.

```javascript
for (var i=0; i<xrefs.length; i++) {
    const xref = xrefs[i];
    tippyHover(xref, undefined, function(instance) {
        instance.disable();
        let url = xref.getAttribute('href');
        let hash = undefined;
        if (url.startsWith('#')) {
            hash = url;
        } else {
            try {
                hash = new URL(url).hash;
            } catch {}
        }
        if (hash) {
            const id = hash.replace(/^#\/?/, "");
            const note = window.document.getElementById(id);
            if (note !== null) {
                try {
                    const html = processXRef(id, note.cloneNode(true));
                    instance.setContent(html);
                } finally {
                    instance.enable();
                    instance.show();
                }
            } else {
                // See if we can fetch this
                fetch(url.split('#')[0])
                    .then(res => res.text())
                    .then(html => {
                        const parser = new DOMParser();
                        const htmlDoc = parser.parseFromString(html, "text/html");
                        const note = htmlDoc.getElementById(id);
                        if (note !== null) {
                            const html = processXRef(id, note);
                            instance.setContent(html);
                        }
                    }).finally(() => {
                        instance.enable();
                        instance.show();
                    });
            }
        } else {
            // See if we can fetch a full url (with no hash to target)
            // This is a special case and we should probably do some content thinning / targeting
            fetch(url)
                .then(res => res.text())
                .then(html => {
                    const parser = new DOMParser();
                    const htmlDoc = parser.parseFromString(html, "text/html");
                    const note = htmlDoc.querySelector('main.content');
                    if (note !== null) {
                        // This should only happen for chapter cross references
                        // (since there is no id in the URL)
                        // remove the first header if (note.children.length > 0 && note.children[0].tagName === "HEADER") {
                        note.children[0].remove();
                        }
                        const html = processXRef(null, note);
                        instance.setContent(html);
                    }
                }).finally(() => {
                    instance.enable();
                    instance.show();
                });
        }
    }, function(instance) {
    });
}
```

--------------------------------

### List Available Tools for an Agent

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/tools.html

Lists all the tools that have been registered with an agent. This is useful for verifying which tools are currently available for the agent to use.

```r
weather_agent$list_tools()
```

--------------------------------

### Initialize Tippy.js Tooltips for Footnotes

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/NEWS.html

Configures Tippy.js to display footnote content in tooltips when hovering over footnote references. Ensure Tippy.js is loaded before this script.

```javascript
const tippyHover = (el, contentFn, onShowFn, onHideFn) => {
  const config = {
    allowHTML: true,
    content: '<span>A Tooltip</span>',
    trigger: 'mouseenter focus',
    delay: [200, 0],
    offset: [0, 5],
    dragDelay: 100,
    maxWidth: 350,
    animation: 'scale',
    arrow: false,
    appendTo: function(el) { return el.parentElement; },
    interactive: true,
    interactiveBorder: 10,
    theme: 'quarto',
    placement: 'bottom-start',
  };
  if (contentFn) { config.content = contentFn; }
  if (onTriggerFn) { config.onTrigger = onTriggerFn; }
  if (onUntriggerFn) { config.onUntrigger = onUntriggerFn; }
  window.tippy(el, config);
}
```

--------------------------------

### Initialize Annotation and Citation Handlers

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/tools.html

Event listeners for clicking annotation targets and logic to populate bibliographic reference popups.

```javascript
window.addEventListener( "resize", throttle(() => { elRect = undefined; if (selectedAnnoteEl) { selectCodeLines(selectedAnnoteEl); } }, 10) ); // Attach click handler to the DT const annoteDls = window.document.querySelectorAll('dt[data-target-cell]'); for (const annoteDlNode of annoteDls) { annoteDlNode.addEventListener('click', (event) => { const clickedEl = event.target; if (clickedEl !== selectedAnnoteEl) { unselectCodeLines(); const activeEl = window.document.querySelector('dt[data-target-cell].code-annotation-active'); if (activeEl) { activeEl.classList.remove('code-annotation-active'); } selectCodeLines(clickedEl); clickedEl.classList.add('code-annotation-active'); } else { // Unselect the line unselectCodeLines(); clickedEl.classList.remove('code-annotation-active'); } }); } const findCites = (el) => { const parentEl = el.parentElement; if (parentEl) { const cites = parentEl.dataset.cites; if (cites) { return { el, cites: cites.split(' ') }; } else { return findCites(el.parentElement) } } else { return undefined; } }; var bibliorefs = window.document.querySelectorAll('a[role="doc-biblioref"]'); for (var i=0; i<bibliorefs.length; i++) { const ref = bibliorefs[i]; const citeInfo = findCites(ref); if (citeInfo) { tippyHover(citeInfo.el, function() { var popup = window.document.createElement('div'); citeInfo.cites.forEach(function(cite) { var citeDiv = window.document.createElement('div'); citeDiv.classList.add('hanging-indent'); citeDiv.classList.add('csl-entry'); var biblioDiv = window.document.getElementById('ref-' + cite); if (biblioDiv) { citeDiv.innerHTML = biblioDiv.innerHTML; } popup.appendChild(citeDiv); }); return popup.innerHTML; }); } } });
```

--------------------------------

### Initialize Tooltips with Hover Effect

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/leadagent.html

Initializes tooltips for elements with the 'tippyHover' function. Ensure the 'tippy' library is loaded.

```javascript
window.tippy(el, config); }
```

--------------------------------

### Visualize Agent Plan

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/leadagent.html

Call visualize_plan() after generating a plan to view the agent orchestration as a diagram. This requires the DiagrammeR package. If no plan exists, it prompts to generate one first.

```R
lead_agent$visualize_plan()
```

--------------------------------

### Set Agent Budget and Policy

Source: https://context7.com/feddelegrand7/mini007/llms.txt

Demonstrates setting a spending budget for an agent and configuring its behavior when the budget is approached or exceeded. Tracks usage statistics.

```r
openai_4_1_mini <- ellmer::chat(
  name = "openai/gpt-4.1-mini",
  api_key = Sys.getenv("OPENAI_API_KEY"),
  echo = "none"
)

agent <- Agent$new(
  name = "cost_conscious_assistant",
  instruction = "Answer succinctly.",
  llm_object = openai_4_1_mini
)

# Set a 5 USD budget
agent$set_budget(5)
# Budget successfully set to 5$
# Budget policy: on_exceed='abort', warn_at=0.8

# Configure budget policy:
# - on_exceed: "abort" (stop), "warn" (continue with warning), or "ask" (interactive)
# - warn_at: fraction (0-1) of budget to trigger warning
agent$set_budget_policy(on_exceed = "ask", warn_at = 0.9)
# Budget policy set: on_exceed='ask', warn_at=0.9

# Check usage statistics
stats <- agent$get_usage_stats()
# $estimated_cost
# [1] 0.0012
# $budget
# [1] 5
# $budget_remaining
# [1] 4.9988
```

--------------------------------

### Initialize Tippy Tooltips for Cross-References

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/index.html

Iterates through cross-reference links and attaches tooltip behavior, including fetching remote content if the target is not in the current document.

```javascript
for (var i=0; i<xrefs.length; i++) { const xref = xrefs[i]; tippyHover(xref, undefined, function(instance) { instance.disable(); let url = xref.getAttribute('href'); let hash = undefined; if (url.startsWith('#')) { hash = url; } else { try { hash = new URL(url).hash; } catch {} } if (hash) { const id = hash.replace(/^#\/?/, ""); const note = window.document.getElementById(id); if (note !== null) { try { const html = processXRef(id, note.cloneNode(true)); instance.setContent(html); } finally { instance.enable(); instance.show(); } } else { // See if we can fetch this fetch(url.split('#')[0]) .then(res => res.text()) .then(html => { const parser = new DOMParser(); const htmlDoc = parser.parseFromString(html, "text/html"); const note = htmlDoc.getElementById(id); if (note !== null) { const html = processXRef(id, note); instance.setContent(html); } }).finally(() => { instance.enable(); instance.show(); }); } } else { // See if we can fetch a full url (with no hash to target) // This is a special case and we should probably do some content thinning / targeting fetch(url) .then(res => res.text()) .then(html => { const parser = new DOMParser(); const htmlDoc = parser.parseFromString(html, "text/html"); const note = htmlDoc.querySelector('main.content'); if (note !== null) { // This should only happen for chapter cross references // (since there is no id in the URL) // remove the first header if (note.children.length > 0 && note.children[0].tagName === "HEADER") { note.children[0].remove(); } const html = processXRef(null, note); instance.setContent(html); } }).finally(() => { instance.enable(); instance.show(); }); } }, function(instance) { }); }
```

--------------------------------

### Configure Navigation Bar Settings

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/leadagent.html

Defines the JSON configuration for the project's navigation bar, including search settings and keyboard shortcuts.

```json
{ "location": "navbar", "copy-button": false, "collapse-after": 3, "panel-placement": "end", "type": "overlay", "limit": 50, "keyboard-shortcut": [ "f", "/", "s" ], "show-item-context": false, "language": { "search-no-results-text": "No results", "search-matching-documents-text": "matching documents", "search-copy-link-title": "Copy link to search", "search-hide-matches-text": "Hide additional matches", "search-more-match-text": "more match in this document", "search-more-matches-text": "more matches in this document", "search-clear-button-title": "Clear", "search-text-placeholder": "", "search-submit-button-title": "Submit", "search-label": "Search" } }
```

--------------------------------

### Initialize Tooltips for Bibliographic References

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/index.html

Iterates through all elements with the role 'doc-biblioref' and uses the 'findCites' function to locate associated citation data. For each reference, it initializes a 'tippyHover' tooltip that displays the full citation content fetched from corresponding 'ref-' elements.

```javascript
var bibliorefs = window.document.querySelectorAll('a[role="doc-biblioref"]');
for (var i=0; i<bibliorefs.length; i++) {
  const ref = bibliorefs[i];
  const citeInfo = findCites(ref);
  if (citeInfo) {
    tippyHover(citeInfo.el, function() {
      var popup = window.document.createElement('div');
      citeInfo.cites.forEach(function(cite) {
        var citeDiv = window.document.createElement('div');
        citeDiv.classList.add('hanging-indent');
        citeDiv.classList.add('csl-entry');
        var biblioDiv = window.document.getElementById('ref-' + cite);
        if (biblioDiv) {
          citeDiv.innerHTML = biblioDiv.innerHTML;
        }
        popup.appendChild(citeDiv);
      });
      return popup.innerHTML;
    });
  }
}
```

--------------------------------

### Tippy.js Tooltip Initialization

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/agents.html

Configures and initializes Tippy.js for creating tooltips on elements. It allows for custom content, trigger functions, and various placement and theming options.

```javascript
function tippyHover(el, contentFn, onTriggerFn, onUntriggerFn) {
  const config = {
    allowHTML: true,
    maxWidth: 500,
    delay: 100,
    arrow: false,
    appendTo: function(el) {
      return el.parentElement;
    },
    interactive: true,
    interactiveBorder: 10,
    theme: 'quarto',
    placement: 'bottom-start',
  };
  if (contentFn) {
    config.content = contentFn;
  }
  if (onTriggerFn) {
    config.onTrigger = onTriggerFn;
  }
  if (onUntriggerFn) {
    config.onUntrigger = onUntriggerFn;
  }
  window.tippy(el, config);
}

const noterefs = window.document.querySelectorAll('a[role="doc-noteref"]');
for (var i=0; i<noterefs.length; i++) {
  const ref = noterefs[i];
  tippyHover(ref, function() {
    // use id or data attribute instead here
    let href = ref.getAttribute('data-footnote-href') || ref.getAttribute('href');
    try {
      href = new URL(href).hash;
    } catch {}
    const id = href.replace(/^#\/?/, "");
    const n
```

--------------------------------

### Initialize Color Mode

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/NEWS.html

Initializes the color mode based on the author's preference or system settings. It disables default stylesheets and sets the initial color mode. If the alternate sentinel is present, it switches to dark mode.

```javascript
document.querySelector('link#quarto-text-highlighting-styles.quarto-color-scheme-extra').rel = 'disabled-stylesheet';
document.querySelector('link#quarto-bootstrap.quarto-color-scheme-extra').rel = 'disabled-stylesheet';
let localAlternateSentinel = darkModeDefault ? 'alternate' : 'default';
if (hasAlternateSentinel()) {
  toggleColorMode(true);
} else {
  toggleColorMode(false);
}
```

--------------------------------

### Tippy.js Hover Tooltip Configuration

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/leadagent.html

Sets up configuration options for Tippy.js tooltips, including HTML content, maximum width, delay, arrow visibility, append target, interactivity, theme, and placement.

```javascript
function tippyHover(el, contentFn, onTriggerFn, onUntriggerFn) {
 const config = {
 allowHTML: true,
 maxWidth: 500,
 delay: 100,
 arrow: false,
 appendTo: function(el) {
 return el.parentElement;
 },
 interactive: true,
 interactiveBorder: 10,
 theme: 'quarto',
 placement: 'bottom-start',
 };
 if (contentFn) {
 config.content = contentFn;
 }
 if (onTriggerFn) {
 config.onTrigger = onTriggerFn;
 }
 if (onUntriggerFn) {
 config.onUntrigger = onUntriggerFn;
 }
 wi
```

--------------------------------

### Generate and Register Tools Automatically

Source: https://context7.com/feddelegrand7/mini007/llms.txt

Allows the agent to create and register its own tools based on a natural language description.

```r
openai_4_1_mini <- ellmer::chat(
  name = "openai/gpt-4.1-mini",
  api_key = Sys.getenv("OPENAI_API_KEY"),
  echo = "none"
)

agent <- Agent$new(
  name = "tool_creator",
  instruction = "You are a helpful assistant.",
  llm_object = openai_4_1_mini
)

# Generate a tool from description
agent$generate_and_register_tool(
  description = "Create a tool that calculates the factorial of a number"
)
# ── Generating tool from description ──
# Description: Create a tool that calculates the factorial of a number
# ── The following tool will be registered ──
# factorial_function <- function(n) { ... }
# Tool successfully generated and registered

# The tool is now available
agent$list_tools()
# [1] "factorial"

# Use the tool
agent$invoke("What is the factorial of 5?")
# [1] "The factorial of 5 is 120."
```

--------------------------------

### LeadAgent$invoke()

Source: https://context7.com/feddelegrand7/mini007/llms.txt

Executes a full prompt pipeline by decomposing tasks, delegating to registered agents, and chaining results sequentially.

```APIDOC
## POST LeadAgent$invoke()

### Description
Executes the full prompt pipeline: decomposition, delegation to agents, and invocation with result chaining. Returns the final response from the last agent in the sequence.

### Parameters
#### Request Body
- **prompt** (string) - Required - The task or prompt to be processed by the agent workflow.

### Response
#### Success Response (200)
- **response** (string) - The final output generated by the last agent in the sequence.
```

--------------------------------

### Set Human In The Loop (HITL) steps

Source: https://context7.com/feddelegrand7/mini007/llms.txt

Configures manual intervention points in the workflow to allow review, editing, or termination of execution.

```r
openai_4_1_mini <- ellmer::chat(
  name = "openai/gpt-4.1-mini",
  api_key = Sys.getenv("OPENAI_API_KEY"),
  echo = "none"
)

researcher <- Agent$new(
  name = "researcher",
  instruction = "You are a research assistant.",
  llm_object = openai_4_1_mini
)

summarizer <- Agent$new(
  name = "summarizer",
  instruction = "You summarize text into bullet points.",
  llm_object = openai_4_1_mini
)

translator <- Agent$new(
  name = "translator",
  instruction = "You translate English to German.",
  llm_object = openai_4_1_mini
)

lead_agent <- LeadAgent$new(name = "Leader", llm_object = openai_4_1_mini)
lead_agent$register_agents(c(researcher, summarizer, translator))

# Set HITL at step 1 (after researcher)
lead_agent$set_hitl(steps = 1)
# HITL successfully set at step(s) 1.

# Can set multiple steps
lead_agent$set_hitl(steps = c(1, 2))
# HITL successfully set at step(s) 1, 2.

# When invoke() runs, it pauses at HITL steps with options:
# [1] Continue with this response
# [2] Edit the response
# [3] Stop the workflow
lead_agent$invoke("Research and translate information about Algeria")
# ── HITL Step 1 ──
# Agent: researcher
# Prompt: Research information about Algeria
# Response: Algeria is the largest country in Africa...
# [1] Continue with this response ---
# [2] Edit the response ---
# [3] Stop the workflow ---
# Your choice [1/2/3]:
```

--------------------------------

### System Instruction Management

Source: https://github.com/feddelegrand7/mini007/blob/main/docs/agents.html

Method to update the agent's system prompt during an active session.

```APIDOC
## update_instruction(new_instruction)

### Description
Updates the agent's system prompt mid-session, affecting both the first system message and the underlying ellmer system prompt.

### Parameters
- **new_instruction** (string) - Required - The new system prompt text.
```