ashishpatel26/500-AI-Agents-Projects

This repository serves as a curated collection of standalone AI agent templates and an educational course for building multi-agent systems using frameworks like CrewAI, LangChain, and LlamaIndex. It provides a React-based web interface to browse and visualize these agents, alongside governance standards and dependency manifests to support development and reproducibility.

This update introduces a collection of new, standalone AI agent templates under the \agents/\ directory, each with its own entry point, configuration, and documentation. The new agents include: a Web Research Agent (LangGraph + Tavily) that synthesizes search results into reports; a Code Review Agent (LangChain) for analyzing code for bugs and security issues; a PDF Q&A Agent (LlamaIndex) for interactive document chat; a SQL Query Agent (LangChain) that answers natural language questions against SQLite databases in read-only mode by default; an Email Drafting Agent (CrewAI) using a two-agent system for professional email composition; a News Summarizer Agent (LangChain) that fetches and summarizes news articles; a GitHub Issue Triager (LangGraph) that automatically assigns severity and labels to issues; a Data Analysis Agent (LangChain + pandas) for querying CSV/Excel files with natural language; a Resume Parser Agent (LangChain) that extracts structured candidate data and scores fit against job descriptions; a Meeting Notes Agent (LangChain) that converts transcripts into structured markdown; and a Stock Research Agent (LangChain). Each agent includes a \.env.example\ file for API key configuration, a \metadata.yaml\ for framework and LLM details, and a \README.md\ with setup and usage instructions.

agents · high confidence

Added a new educational course in the \crewai\_mcp\_course\ directory that teaches beginner developers how to integrate CrewAI agents with FastMCP servers. The course includes three progressive lessons: Lesson 1 covers basic agent setup and task execution; Lesson 2 demonstrates multi-agent crews with custom tools (word counter, text formatter) and sequential processes; and Lesson 3 shows how to mirror FastMCP server tools as CrewAI wrappers for complex workflows, including a standalone \mcp\_server.py\ example that exposes tools like \get\_datetime\, \prioritize\_tasks\, \format\_as\_json\, and \calculate\_project\_metrics\ via a FastMCP server on localhost:8000. Each lesson provides complete Python scripts, environment variable examples, and requirements files.

_crewai\_mcp\course · high confidence

A new React application has been added to the web directory, providing a dark-themed, single-page interface for browsing the repository's 500+ AI agent projects. The app features a navigation shell with hash-based routing for sections including the CrewAI MCP course, runnable agents, use cases, and frameworks. It dynamically parses Markdown files from the repository root, agents, and course directories to populate a searchable catalog, displaying statistics and an interactive graph visualization on the home page. The site is configured for deployment on GitHub Pages via Vite.

web · high confidence

The repository now enforces a structured contribution workflow and legal compliance. Contributors must sign off on every commit using the Developer Certificate of Origin (DCO) via the \-s\ flag, a requirement enforced by CI checks. A comprehensive CONTRIBUTION.md defines project folder structures, metadata schemas, and reproducibility standards. Additionally, standard governance files have been added: an MIT LICENSE, a CODE\_OF\_CONDUCT.md, a SECURITY.md outlining vulnerability reporting procedures, and a .markdownlint-cli2.jsonc configuration to enforce documentation quality on key files.

(repo-wide) · high confidence

The broken star-history.com image in the README has been replaced by a locally generated SVG chart. A new script (scripts/star-history.mjs) fetches stargazer data from the GitHub API and renders the history curve directly, ensuring the chart remains visible and accurate without relying on the external service.

scripts · high confidence

This change introduces requirements.txt files for 21 new AI agent modules (ranging from web research and code review to PII sanitization) and a new React-based web interface. The agents rely primarily on the LangChain ecosystem (v0.3.0), CrewAI (v0.80.0), and LlamaIndex (v0.11.0), with specific integrations for OpenAI, Tavily, and FAISS. The web interface is built with React 18.3.1 and Vite 6.0.1, utilizing Lucide React for icons.

(dependencies) · high confidence

Written by watchdog.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

• First survey — no prior run to compare against. CAI 56.

• Code Health 93
• Architecture 79
• Maturity 72
• Readiness 35
• Security 71
• Accessibility 82

• 66 commits — 62 feature/other, 4 fixes

• (root) — 45 commits
• (repo) — 13 commits
• .github/workflows — 3 commits
• agents/01-web-research-agent — 1 commit
• agents/16-documentation-writer — 1 commit
• agents/21-pii-sanitization-agent — 1 commit
• images/industry_usecase1.png — 1 commit
• web/src — 1 commit

• fix: Merge pull request #14 from ashishpatel26/copilot/fix-10d2ad23-53e7-4053-9882-918cfdf5017d
• fix: Merge pull request #151 from ashishpatel26/fix/docs-and-links
• fix: fix: replace broken star-history.com chart with a self-generated one (#156)
• fix: fix: stop duplicating class methods in documentation-writer agent (#137)
• change: 59 new agent usecase added
• change: ADD cybersecurity use case.md
• change: Add Agent Wallet SDK to Finance use cases
• change: Add CONTRIBUTION.md and LICENSE (MIT)
• change: Add Citadel — open-source agent orchestration harness
• change: Add Cybersecurity Agent use case
• change: Add MIT License to the project
• change: Add/update CONTRIBUTION.md and add LICENSE (MIT)
• change: Agno Agent Usecases added
• change: All framework code added
• change: Create jekyll-gh-pages.yml
• change: CrewAI Broken Links Updated
• change: CrewAI UseCases Added
• change: Frameworkwise usecase added
• change: Initial commit
• change: Initial plan
• …and 46 more

Written by watchdog.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

ashishpatel26/500-AI-Agents-Projects was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point a surveyor at a repository you know and see whether you agree with it.