The OWASP Social OSINT Agent is an autonomous tool aimed at enhancing open-source intelligence (OSINT) investigations across various social media platforms. It pulls data from sites like Twitter, Reddit, Bluesky, GitHub, Hacker News, and Mastodon, using large language models (LLMs) via OpenAI-compatible APIs. The agent generates structured analytical reports, transforming disparate public information into actionable insights. Key features include multi-platform data collection and high-fidelity OSINT extraction, which captures not just text but also platform-specific elements, such as GitHub commit messages and Reddit user karma. The agent employs a two-phase analysis approach, first gathering all textual data and media before moving to image analysis, optimizing efficiency. It can compare profiles across platforms and handles errors gracefully, ensuring that individual failures don't disrupt the entire process. The system includes robust caching to minimize API calls and accelerate subsequent analyses, with data cached for 24 hours. Users can run the agent interactively or in batch mode, using JSON inputs for automated workflows. Security measures are in place, including environment variable configuration for managing sensitive information. The workflow is visually represented in an interactive flowchart, detailing each step from setup to output generation, providing transparency into the agent's decision-making process.