Search Results for "ai code agent"

The claude-code-security-review repository implements a GitHub Action that uses Claude (via the Anthropic API) to perform semantic security audits of code changes in pull requests. Rather than relying purely on pattern matching or static analysis, this action feeds diffs and surrounding context to Claude to reason about potential vulnerabilities (e.g. injection, misconfigurations, secrets exposure, etc). When a PR is opened, the action analyzes only the changed files (diff-aware scanning),...

PentestGPT is an AI-powered autonomous penetration testing agent designed to perform intelligent, end-to-end security assessments using large language models. Published at USENIX Security 2024, it combines advanced reasoning with an agentic workflow to automate tasks traditionally handled by human pentesters. The platform supports multiple penetration testing categories, including web security, cryptography, reversing, forensics, privilege escalation, and binary exploitation. ...

...It begins with a multi-phase reconnaissance engine that maps the entire attack surface of a target, collecting information such as subdomains, open ports, services, and potential vulnerabilities. RedAmon then uses an AI agent orchestrator to analyze this data, select appropriate tools, and perform exploitation steps such as credential brute forcing or CVE-based attacks. All discovered assets, relationships, and vulnerabilities are stored in a Neo4j knowledge graph, allowing the system to reason about the environment and make informed decisions during the attack process.

Security is a subject of major concern in mobile agents. This project provides a security framework that allows the protection of mobile agents from third parties. A self-protected approach, where the agents themselves manage the process, is used.

TACS (Trust Ant Colony System) is a Trust model for P2P, Ad-hoc and Wireless Sensor networks (also valid for multi-agent systems) based on the bio-inspired algorithm ACS (Ant Colony System).