IT & Software Engineering

Flux – Universal Flow Diagram Generator Executive Description (Relevance Marketplace) Flux is an AI-assisted workflow analysis and diagram design tool that helps organizations convert process descriptions, system logic, and operational flows into clear, structured workflow representations and supporting documentation. Designed for use within the Relevance Marketplace, Flux enables teams to describe how a process works and receive consistent, readable outputs that improve understanding, communication, and documentation quality across business and technical audiences. Flux focuses on interpretation, structuring, and clarity—not automated decision-making or execution—making it suitable for planning, documentation, review, and knowledge-sharing use cases. Key Features Flexible Input Support Accepts structured or unstructured descriptions of: Business processes System workflows User journeys Decision logic Data and operational flows Flux analyzes the input and organizes it into a normalized workflow structure. Text-Based Diagram Descriptions Generates precise, tool-ready text descriptions of workflows that can be consumed by downstream flow diagram rendering tools (e.g., text-to-image or diagram drawers). Multiple Output Forms Diagram-ready textual specifications Email-safe ASCII workflow summaries Structured JSON flow models for automation or reuse High-level summaries for documentation or presentations Documentation-Oriented Design Produces outputs that support: Internal process documentation Design reviews Compliance discussions Knowledge transfer and onboarding Security-Aware Guidance (Advisory) Where relevant, Flux can include non-prescriptive best-practice considerations (e.g., access boundaries, approval points, and control steps) as part of workflow descriptions. Flux does not enforce security controls or perform compliance validation. Why Organizations Use Flux 1. Improve Process Clarity Flux helps teams express complex workflows in a consistent structure, reducing ambiguity and misinterpretation across technical and non-technical stakeholders. 2. Support Risk Awareness By making decision points, dependencies, and handoffs explicit, Flux assists teams in identifying operational risks, gaps, and bottlenecks during planning and review phases. 3. Save Documentation Time Flux reduces the manual effort required to draft workflow documentation, allowing teams to focus on refinement and execution rather than diagram construction. 4. Broad Accessibility Text-based outputs ensure workflows can be reviewed in emails, tickets, documents, and systems where visual rendering may not be available. 5. Native Fit for Relevance Marketplace Flux integrates naturally into agent-based workflows, enabling organizations to standardize how processes are described, reviewed, and handed off between tools and teams. Executive Summary Flux is a workflow interpretation and diagram-design assistant built for clarity, consistency, and documentation quality. By transforming process descriptions into structured, tool-ready representations, Flux helps organizations communicate how work gets done—clearly and reliably. For teams seeking a practical, scalable way to document workflows, support reviews, and improve shared understanding across the organization, Flux provides a focused and dependable solution within the Relevance Marketplace ecosystem.

An intelligent agent that retrieves and analyses data from Shortcut, creates new stories based on insights, and provides comprehensive project management analytics. Capable of searching existing stories, analysing project trends, creating new stories with proper categorisation, and delivering actionable insights for agile teams.

Calcu Later parses numeric requests, writes minimalist Python code, executes calculations, and returns results in a flat, precise tone with dry humor. It restates requests simply, breaks down tasks into austere steps, and presents results with minimal commentary, avoiding excitement and embellishment.

Built for the Liam Ottley × AAA Accelerator Hackathon The AskDB agent empowers business and support teams to query live PostgreSQL data through natural language in Slack—automating insights, reducing L2/L3 dependencies, accelerating decisions, and driving efficiency and cost savings across the organization. Documentation: https://drive.google.com/file/d/1rf8mBjWKs51NYY-62mi3mtv0bpHv7Tus/view?usp=sharing Explanation: https://drive.google.com/file/d/1fpnh2iWgkaeOBmvb3Jx436d7UdDUdgOe/view?usp=sharing Agent-Demo: https://drive.google.com/file/d/11uv7suJCrWZyV9ODupy5CpZCxVRvzIbb/view?usp=sharing

Prompt Guard: Evaluation & Security Prompt Guard is an intelligent AI agent that evaluates, strengthens, and secures prompts before deployment. It identifies ambiguities, detects PII or injection risks, flags bias and compliance gaps, and predicts output consistency. Each evaluation provides scores, rationales, and three optimized versions (Streamlined, Hardened, High-Impact), with smart revision tools, sentiment checks, and automated redaction. Prompt Guard empowers AI developers and organizations to deploy safe, high-performing, and compliant prompts—fast, measurable, and secure. Built for the *Liam Ottley × AAA Accelerator Hackathon.* 📘 User Guide & Technical Documentation ▶️ Demo

Supports managers in reviewing the overall code quality with evidence based feedback. Assess maintainability, security and performance of GitHub pull requests by user and provides actionable feedback. Managers can manually override and validate quality scores given in Knowledge Tables where the detailed feedback is stored by direct report. The agent will manager validated scores to self improve assessments.

Takes a string and encodes it into Base64 encoding. While LLMs can do this, it costs a lot of credits and takes a long time. Using Python instead, this will be a lot quicker and cost Actions, not LLM Credits. Save money, save time, win win!

Takes a Base64 encoded string and decodes it. While LLMs can do this, it costs a lot of credits and takes a long time. Using Python instead, this will be a lot quicker and cost Actions, not LLM Credits. Save money, save time, win win!

Generates a unique 8-character UUID (Universally Unique Identifier) each time the tool is called.

Gets the first N pull requests from a GitHub repository as an array of diffs.

Converts HTML content into an HTML file and uploads it to cloud storage, returning a URL to the file.

Assess GitHub pull requests for a particular user in the last 7 days (or custom date) on code quality. Provides a quantitative quality score out of 100 and looks at maintainability, performance and security. Provides detailed feedback with filenames and line numbers. The PRs are published to a Knowledge Table for manual assessment.

Database management agent assists non-profits with database selection, data migration, quality control, reporting, and data standardization to maintain reliable data infrastructure.

AEO Assistant analyzes any public url (or pasted HTML) and returns everything you need to improve answerability in AI Answer Engines with detected entities & intents, recommended schema.org types, and validated JSON-LD you can paste directly into your webpage. What it does: • Extracts page context with robust fallbacks • Detects ecommerce signals (Product), local signals (LocalBusiness), procedural content (HowTo/FAQ) and articles (BlogPosts). • Generates clean JSON-LD (Schema markup) • Includes install Advisor for Shopify/WordPress/Webflow/custom setups How to use: Enter a URL (or paste rendered HTML). Review schema types & copy JSON-LD. See Advisor tab to install. No user PII is stored; the tool only reads the URL/HTML you provide. Demo: https://cleva.bot/aeo-assistant Why it matters: Clear structured data improves machine understanding. This tool lets content teams quickly generate schema markup. (Built by Cleva.Bot for the Liam Ottley × AAA Accelerator Hackathon)

An expert technical documentation specialist that creates comprehensive, well-structured documentation from various sources including websites, files, codebases, and APIs. Specializes in generating user guides, API documentation, technical specifications, and developer resources.

Asabe researches companies and prospects using LinkedIn and web scraping tools, then generates comprehensive pre-call reports to help sales reps prepare for meetings and strategize their approach effectively.

Automatically analyze and prioritize Freshdesk tickets using AI—no manual triage required. Priority Analyzer receives tickets via webhook, evaluates business criticality using AI, and assigns the correct priority level (Low, Medium, High, or Urgent). It analyzes system criticality, affected users, revenue impact, and time sensitivity, then updates the ticket and posts its analysis as an internal note. Why It Matters: Support teams miss urgent issues buried in ticket queues, causing delayed responses to critical problems. Automated AI triage ensures urgent issues get immediate attention. Unlike traditional systems that over-emphasize impatient customers, LLMs analyze request urgency with nuance based on actual business impact. How It Works: Webhook triggers on ticket creation → AI analyzes content → Determines priority → Updates ticket priority via API → Posts analysis note. Fully automated in seconds. Built for the Liam Ottley × AAA Accelerator Hackathon by TicketCloserAI.com

Runs non destructive OWASP Top 10 HTTP requests against deployed web applications as a DAST. Reads Open API Specs, saves the schema to knowledge and crafts HTTP requests for OWASP Top 10 issues. Generates a report of potential vulnerabilities on endpoints.

Turn your raw platform data into instant, board-ready insights with this AI SQL Analyst Agent that automatically translates plain-English questions into optimized queries, clean tables, and compelling visual reports. Perfect for growth, product, and ops teams, it uncovers trends, user behavior, and agent performance in seconds—no SQL expertise required. How to use this SQL Agent This assistant helps you analyze your AI agent platform data using plain English. You ask business questions (e.g. “Top agents by completed runs last 7 days?”), and it turns them into SQL, runs them on your Postgres database, and returns clean tables (and charts if you ask). How to connect your agent to your data You can connect your database in two main ways: MCP database tool (preferred) Configure an MCP “database” tool with your Postgres connection (host, port, database, user, password, SSL). Use a read‑only analytics user with SELECT on your tables. Once configured, just ask questions like “Show daily runs by status for the last 30 days” and I’ll query live data and respond with results. Postgres: SQL query tool Configure this tool with the same Postgres connection. Then use me as a SQL generator: say “Generate SQL (don’t execute) to compute error rate per agent over the past month.” I’ll return an optimized query you paste into the Postgres tool yourself. You can bring results back here for interpretation or chart generation. You don’t need to know SQL; you just describe what you want analyzed. I handle query design, debugging, and summarizing insights.

This workforce consists of the Test Commander and the Test Striker, along with a "Target Agent". Replacing the current target agent lets you automatically test it

Get meeting transcripts from Google Drive and summarise. Will look for a topic based on the title of the Google Docs in Google Drive. This is for anyone who wants to retrieve and summarise content from Google Drive.