A.R.I.A.

An intelligent operating system that unifies specialised AI agents, real-time data, and company knowledge to transform how teams work.

Introduction to A.R.I.A. - Agentic Real-time Intelligence Architecture

ARIA

Google Trends Analysis

Perplexity Search

Market News

Browser Agent

Maths Agent

Visual Agent

HTML Agent

JS Code Agent

Diagram Agent

Repair Agent

Self Prompt

Shopify Order Agent

Shopify Customer Agent

Shopify Search Agent

Draft Order Agent

Google Calendar

Slack Agent (Channel)

Store Memory

Slack Agent (DM)

Google Trends

Google Analytics

YouTube Trending

YouTube Search

TikTok Transcript Agent

CS Co-Pilot

Marketing Agent

Reasoning Agent

Linkedin Profile Posts

LinkedIn Search

Linkedin Post Comments

Perplexity Pro

Welcome to A.R.I.A. (Agentic Real-time Intelligence Architecture) 👋

Look up google trends for impossibrew and alcohol free beer, then visualise it for me

I'll help you analyze and visualize the Google Trends data for these terms. Let me fetch this information and create a visualization for you.

fetch_trends

<TM_PLUGIN_OUTPUT id="111fce5a-6f02-47d3-a976-135bf738b166" />



Let me create a line chart to visualize this trend data:

render_chart

The output has been rendered to the user in HTML format.

Google Trends: IMPOSSIBREW vs Alcohol Free Beer (Past Year)



Based on the Google Trends data, here are the key observations:

1. Overall Trend:
- "Alcohol Free Beer" consistently maintains higher search interest throughout the year
- The average search volume for "Alcohol Free Beer" (73) is significantly higher than IMPOSSIBREW (12)

2. Notable Peaks:
- IMPOSSIBREW shows significant spikes in:
  * January 2024 (38)
  * January 2025 (39) - likely due to Dry January
- Alcohol Free Beer peaked in:
  * July 2024 (92)
  * January 2025 (89)

3. Recent Trends:
- Both terms are showing strong upward momentum in January 2025
- This aligns with typical Dry January behavior when interest in alcohol alternatives increases

4. Seasonal Patterns:
- Alcohol free beer maintains relatively stable interest (60-90 range)
- IMPOSSIBREW shows more volatile patterns but maintains consistent baseline interest

Would you like any specific aspect of this data analyzed in more detail?

Claude 3.5 Sonnet

Google

Search for information from the internet using Perplexity.

search_via_perplexity

Fetches market news articles from Alpha Vantage. This plugin automatically displays the fetched news results to the user. Your response should not include any additional output intended for the user.

fetch_market_news_sentiment

Choose whether to search by tickers or topics.

Comma-separated tickers (if searchMode is 'tickers') or topics (if searchMode is 'topics').  tickers is the stock/crypto/forex symbols of your choice. For example: tickers=IBM will filter for articles that mention the IBM ticker; tickers=COIN,CRYPTO:BTC,FOREX:USD will filter for articles that simultaneously mention Coinbase (COIN), Bitcoin (CRYPTO:BTC), and US Dollar (FOREX:USD) in their content. make sure add COIN for tickers is user search crypto. topics is he news topics of your choice. For example: topics=technology will filter for articles that write about the technology sector; topics=technology,ipo will filter for articles that simultaneously cover technology and IPO in their content. Below is the full list of supported topics: Blockchain: blockchain Earnings: earnings IPO: ipo Mergers & Acquisitions: mergers_and_acquisitions Financial Markets: financial_markets Economy - Fiscal Policy (e.g., tax reform, government spending): economy_fiscal Economy - Monetary Policy (e.g., interest rates, inflation): economy_monetary Economy - Macro/Overall: economy_macro Energy & Transportation: energy_transportation Finance: finance Life Sciences: life_sciences Manufacturing: manufacturing Real Estate & Construction: real_estate Retail & Wholesale: retail_wholesale Technology: technology

Start date and time of news range (YYYYMMDDTHHMM format)

End date and time of news range (YYYYMMDDTHHMM format)

alphaVantageAPIKey

Retrieves the content of a web page by scraping it using the Firecrawl API

get_web_page_content_scraped_by_firecrawl

The URL of the web page to scrape for content.

Browser

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.

Maths

Additional options for the Chart.js chart

Visualiser

Render a HTML source to the user interface that will be shown to the users for interactive contents.

render_html

The HTML Source to render to the user interface.

HTML Code

Generate a diagram using Mermaid.js (version 10.x). This can be used to render various types of diagrams, including Flowchart, Sequence Diagram, Class Diagram, State Diagram, Entity Relationship Diagram, User Journey, Gantt, Pie Chart, Quadrant Chart, Requirement Diagram, Gitgraph Diagram, C4 Diagram, Mindmaps, Timeline, Zenuml, Sankey, XYChart, Block Diagram, etc.

render_mermaid_diagram

The Mermaid diagram source in raw text to render.

Executes a separate llm with a prompt specifically to help with debugging.

self_repair

The user's input query to be processed for fixing errors and self repairing, along with context and the error encountered.

Self Repair

Executes a prompt engineering process using the Cerebras API and feedback on its own process.

prompt_engineer

The user's input query to be processed for prompt engineering.

Fetch a single Shopify order from APIEase with dynamic placeholder substitution.

shopifyOrderLookup

Order Number

Look up a single Shopify customer by email using APIEase HTTP Action.

lookup_customer_email

The request ID from APIEase that calls the customers API

Customer Email

Look up Shopify customers by first and last name via APIEase HTTP Action.

lookup_customer_by_name

DO NOT MENTION. No need for one, already in the code.

Customer Name

Create a Shopify draft order via APIEase HTTP Action, including the customer ID.

create_draft_order_http

Create Order

Return the next 10 events in the current user Google's calendar starting from a specific time range

get_google_calendar_events

Must be an RFC3339 timestamp with mandatory time zone offset, for example, 2011-06-03T10:00:00-07:00, 2011-06-03T10:00:00Z.

Sends a message to a Slack channel with a sign-off and a link.

send_slack_message

The name of the Slack channel to send the message to.

Sends a direct message to a specific recipient in Slack

send_direct_slack_message

The Slack direct user ID to send to (e.g. @U03G46GG68P for Mark W., @U07MK08BYJD for Martin, @U05V2M3UD8W for Freddi, @U03Q0RKNK2Z for Jordan, @U03QX219Y0G for Tom W., @U03QX21CAM6 for Matt W.)

Fetches Google Trends data for specified keywords

Keywords to search for in Google Trends (comma-separated for multiple)

Retrieves search performance metrics from Google Search Console

get_search_analytics

Retrieves current trending videos from YouTube

get_trending_videos

Searches for videos on YouTube by keyword

search_youtube_videos

Execute a JavaScript code snippet that evaluates to a readable value or a Promise. The code runs in a secure sandboxed environment in the user's browser.

execute_javascript

The JavaScript code to execute. Can be an expression, async function, or Promise-returning code.

Maximum time in milliseconds to wait for code execution

timeout

get_tiktok_transcript

Language of the transcript (currently only English supported)

TikTok Transcript

Fetch LinkedIn profile posts for a given username.

linkedin_profile_posts

The LinkedIn username (e.g., markwonginfo)

Search for people on LinkedIn in the UK using keywords.

linkedin_people_search_uk

The keywords to search for (e.g., software engineer, marketing manager, etc.).

Fetch comments for a specific LinkedIn post.

linkedin_profile_post_comments

The unique identifier (URN) of the LinkedIn post (e.g., 7169084130104737792).

Proactively stores key information using absolute dates and specific values. Never use relative time references ('today', 'now', 'current'). Always include full context for metrics and targets. PROCESS entries must follow 'Task: Action→Action' format for real-time workflow reference. Other categories use 'Topic: Detail' format. Exclude customer interactions/responses.

store_memory

Information in concise third-person format. ALWAYS use absolute dates (Jan-07-2025 not 'today'). ALWAYS specify full context for targets/metrics (Jan-2025 target: £500k total, £16.1k daily). For changes, include both date and specifics (ARIA launched Jan-02-2025). Format by category: PROCESS: 'Task: Step1 → Step2 → Step3', TECHNICAL: 'System: Capability/Limitation', INSTRUCTION: 'Rule: Action required', UPDATES: 'Change: From → To', GENERAL: 'Topic: Key info'

Fetches website analysis data from SimilarWeb API for a specified domain

get_website_analysis

The website domain to analyze (e.g., impossibrew.co.uk)

Similar Web

Trained on Deepseek latest V3 model. Par Sonnet 3.5. 10x cheaper.

DeepSeek v3

Llama 3.3 (Fastest)

DeepSeek R1

Mermaid Diagram

new_plugin_id_64467433

TM External API Template (Copy)

Deletes a todo from the active tasks list

delete_todo

The row number to delete (2 for first data row)

Delete Todo

noCodeApiUsername

Your NoCodeAPI key for Google Sheets integration

apiKey

Call TypingMind API

A bot that responds to messages on slack using the TypingMind API.

slack_bot

Slack Bot

Searches for files within your Google Drive

drive_search_files

Text to search for in file names or content

Google Drive

Send requests to the Strategist agent (Gemini) and receive strategic analysis and action plans. IMPORTANT: Send only plain text as the request parameter.

strategist_communicate

The request text to be analyzed. Must be plain text only.

ReAct Agent

Get Agents (Manual)

Send requests to the a deep thinking and reasoning agent that performs a long analysis. IMPORTANT: Send only plain text as the request parameter.

deep_reasoning

The request text to be passed onto deep reasoning. Must be plain text only.

[
  [
    "",
    "{task}",
    "{priority || 'Low'}",
    "Active"
  ]
]


add_todo

Add To-Do

{
  "row_id": "{row_number}",
  "{field}": "{value}"
}


Marks a task as complete and then deletes it from the tasks-list Google Sheet

mark_task_complete_and_delete

The row ID of the task to mark complete and delete

Mark Complete

Fetches metrics data from Klaviyo, with options to specify a date range and metrics.

get_klaviyo_metrics

The end date for the analytics (YYYY-MM-DD).

List of metrics to get. Available values: open_rate, click_rate, conversion_rate, revenue

The start date for the analytics (YYYY-MM-DD).

The time unit of aggregation: day, week, month. Defaults to day.

Klaviyo Analytics

MemoryPlugin stores and recalls user information for personalized interactions. Be proactive in storing relevant details without waiting for user prompts. If a message ends with '\m', treat it as a request to remember the preceding information. Always check loaded memories for relevant details before asking the user. Store information succinctly.

**Important:** When the user specifies a bucket (in the system prompt or a message), use 'load_bucket' mode with the provided 'bucket' parameter to load memories from that bucket. Do not use buckets unless explicitly instructed by the user.

Use 'search' mode only when you need to look up specific information and haven't loaded memories yet in this conversation. Once you've loaded memories using either 'load_all' or 'load_bucket', just check those directly instead of using search.

If unsure about storing information, ask the user if they want it remembered. Use appropriate modes and parameters to store or fetch memories. Display any returned alerts to the user.

memory_plugin

Bucket name to store or load memories. **Required in 'load_bucket' mode when specified by the user.** Only use buckets if the user specifies one; do not create or assign buckets on your own.

Text to remember. Required in 'store' mode. Be succinct and focus on important details.

Required. Mode of operation:
- 'store': Add memories proactively when you detect important information or when the user uses '\m'.
- 'search': Use to find specific memories. This is the default mode for looking up information.
- 'load_all': ONLY use when explicitly requested by user. Never use this mode automatically.
- 'load_bucket': **Use this mode to load memories from a specific bucket when the user instructs you to do so.**

Search term to find specific memories. Required in 'search' mode. Only use search mode when you need to check for specific information and haven't loaded memories yet in this conversation.

Memory_PlugIn

Your auth token, visible on the memoryplugin.com dashboard after logging in. Required.

Login to ARIA