Tag: AI coding agents
‘HalluSquatting’ Compromises AI Coding Agents to Install Malware, Create Botnets
Jeff Burt | | AI coding agents, AI hallucinations, AI security agents, Cursor, devops, GitHub Copilot, Google Gemini, HalluSquatting, indirect prompt injection, LLM, OpenClaw, promptwareHallucinations have been an ongoing problem since OpenAI first introduced its ChatGPT chatbot in November 2022, highlighting generative AI’s tendency to generate plausible but false or misleading information and its inability to ...
GhostApproval Flaw Featuring Decades-Old Feature Found in Six AI Coding Tools
Jeff Burt | | Adversa AI, AI coding agents, AI security risks, Amazon Q Developer, Anthropic Claude Code, Augment, CodeSignal, Cursor, GhostApproval, Google Antigravity, human in the loop coding, OpenAI Codex, SSH keys, symlinks, Unix, Windsurf, WizA security flaw found in six popular AI coding agents can let attackers abuse a decades-old feature in Unix to trick an AI agent into giving them control of a developer’s system ...
Novee Uncovers Cordyceps: The Latest Threat to CI/CD Pipelines
Jeff Burt | | AI coding agents, apache, CI/CD security, Cloudflare, code repositories, Cordyceps, GitHub Actions, google, microsoft, Novee, npm malware, PyPi, Python, supply chain attacks, WizA newly discovered supply chain security flaw is once again putting a spotlight on inherent weaknesses in CI/CD pipelines and the growing interest among cyberthreat actors to exploit them. Security researchers with ...
Mozilla Shows the Danger of Indirect Prompt Injections in AI Coding Agents
Jeff Burt | | AI coding agents, AI security risks, Anthropic Claude Code, Developer Security, exposed credentials, exposed secrets, Futurum Group, indirect prompt injection, leaked API keys, Mozilla 0DIN, shell scriptsA clean GitHub repository that contains no malicious code can launch an attack and fully compromise a developer’s systems by using indirect prompt injections to trick AI-powered coding agents like Anthropic’s Claude ...
AI Coding Costs Could Exceed Developer Salaries, Gartner Warns
Enterprises rapidly adopting AI coding assistants may eventually face an unexpected financial reality: the AI tools could cost more than the developers using them. According to new research from Gartner, spending on ...
Tenet’s ‘Agentjacking’ Attack Turns Sentry Errors Into Code Execution
AI coding agents can create a new code execution risk when they treat externally influenced error data as trusted guidance and have access to command line tools, according to new research from ...
GitHub’s Spec Kit Puts the Spec Back in Software Development
Tom Smith | | AI coding agents, devops automation, GitHub Copilot, GitHub Spec Kit, software governance, Spec-Driven Development (SDD), technical debt, vibe codingGitHub’s open-source Spec Kit aims to eliminate "vibe coding" by prioritizing durable specifications over vague prompts, providing a structured, agent-agnostic workflow for Copilot, Claude, and Gemini ...
The Great Decoupling: Scaling the Outer Loop for the Agentic Era
Nassir Khan | | AI coding agents, AI-SDLC, compliance automation, devsecops, software development lifecycleThe "Inner Loop" of software development—the iterative cycle of writing, building, and debugging code—has just broken the sound barrier. With the emergence of agentic coding tools like Claude Code and GitHub Copilot ...
Cursor’s New SDK Turns AI Coding Agents Into Deployable Infrastructure
Tom Smith | | agent control plane, AI coding agents, autonomous pull requests., CI/CD agent pipelines, Cloud execution sandboxing, Composer 2 model, Cursor SDK, devops automation, MCP integration, Model Context Protocol, Multi-agent workflows, Programmatic code generation, semantic search indexing, software lifecycle engineering, TypeScript SDK for AICursor's TypeScript SDK lets teams invoke AI coding agents programmatically from CI/CD pipelines, with sandboxed VMs, MCP support, and token-based pricing ...
5 Facts About AI Coding Agents from Comprehensive Benchmarking
Graham Neubig | | AI coding agents, benchmarking, developer tools, Large Language Models, software developmentAI coding agents are becoming more capable, but evaluating them is harder than it looks. Most benchmarks focus on a single dimension of agent capabilities; for instance, the popular SWE-Bench benchmark only ...
If it Isn’t Code, it’s Just Advice
David Mytton | | AI coding agents, API contracts, audit trails, automated verification, automation vs. advice, CI/CD, code-native, configuration as management, dashboard problem, infrastructure as code, MCP Server, production safety, reproducibility, software development life cycle, software security, spam prevention, UI-driven configuration, unit testing, vendor integrations, version controlIn the world of AI coding, success hinges on code-native solutions that integrate and verify changes directly in the codebase. Many AI tools fail by relying on external dashboards, lacking the reproducibility ...
Meta Researchers Show AI Agents Can Verify Code Without Running It — and Hit 93% Accuracy
Tom Smith | | Agentic Code Reasoning, AI coding agents, Code Equivalence Verification, Defects4J Benchmark, DevOps Code Review, Execution-Free Reward Signals, Meta Research, Reinforcement Learning Training, RubberDuckBench, Semi-Formal Reasoning, software lifecycle engineering, static analysisMeta's semi-formal reasoning enables AI agents to verify code without executing it, achieving 93% accuracy. Implications for code review and RL training costs ...
