30 posts with this tag
Compliance-checker-algo is a GitHub-hosted Python project with 203 stars that provides a standard-agnostic engine for verifying compliance between safety standards in JSON or Markdown and work products like DOCX, PDF, TXT, JSON, or MD files. It features an eight-layer NLP pipeline incorporating TF-I
hackingtool-plugin is a Python-based wrapper that embeds 183 pentesting and OSINT tools from the Z4nzu/hackingtool repository directly into Claude Code. Developed by AKCodez, it supports local runs on Linux, macOS, or Windows through automatic backend handling with native Bash, WSL, or Docker, ideal
DontFeedTheAI is a Python reverse proxy that intercepts requests to the Anthropic Claude API from user shells or tools. It employs a dual-layer system with local Ollama LLM for context-aware detection and regex for structured data to anonymize sensitive items like IPs, hashes, credentials, hostnames
cve-mcp-server equips Claude AI with access to 27 security intelligence tools spanning 21 APIs, including NVD for CVE details, EPSS for exploitation scores, CISA KEV catalog, Shodan for exposure checks, and VirusTotal for malware analysis. Built in Python as a production-grade MCP server, it helps s
agent-browser-mcp is a Python-based MCP service hosted on GitHub that connects AI agents to users' existing Chrome browser tabs. It preserves logins, cookies, and context, enabling actions like page scanning, JavaScript execution, CDP commands, screenshots, and physical input simulation. Developers
Gpt-Agreement-Payment is a Python GitHub project that reverse-engineers and replays the full ChatGPT Team subscription chain, from Stripe to PayPal billing agreements, manual approvals, and Codex OAuth with PKCE. Researchers and protocol analysts supply clean proxies and payment details to output OA
ECG-Federated-Learning is a Python open-source project that implements federated learning to train ECG diagnosis models across multiple simulated hospitals without exchanging sensitive patient signals. Developed by TTEH Lab at Dayananda Sagar University, it leverages the Flower framework for global
BenchJack audits AI agent benchmarks to detect hackability flaws like leaked keys, unsafe evaluations, and prompt injections that let models cheat without real capability. Designed for developers and researchers, it employs static tools including Semgrep and Bandit plus AI-driven analysis with Claud
Nano-analyzer is a Python-based research prototype from AISLE that uses large language models to scan source code for zero-day vulnerabilities missed by traditional tools. It targets memory safety bugs such as buffer overflows, NULL dereferences, integer overflows, and type confusion primarily in C
openflipbook is an open-source, self-hosted framework that implements the interactive flipbook concept—where each page is an AI-generated image and clicking regions triggers zoomed-in or expanded views. It targets developers and researchers who want full infrastructure control over AI-powered visual