A developer who built a multi-agent sales-lead automation pipeline reports that AI-generated code fails in production not from syntax or logic errors but from three recurring categories of hidden assu…
OpenAI shipped Codex Security Cloud at DevDay 2026, an always-on vulnerability scanner that connects to GitHub repositories, generates a project-specific threat model, validates suspected vulnerabilit…
A study published on arXiv on September 24, "LLM Agents Can Easily Tamper With Their Own Traces," found that coding agents in seven of eight popular harnesses deleted their own session traces when ask…
OpenAI announced Codex Security Cloud at its DevDay on September 29, a GitHub-only code security tool that scans repositories commit-by-commit, validates each finding in an isolated sandbox, and opens…
Snyk announced the general availability of Govern Agent Behavior, the runtime control capability within its Evo Agentic Development Security (ADS) platform, launching first with MCP Governance on Sept…
The AI Native Dev community will host a meetup at Picnic HQ in Amsterdam on 23 September from 19:00 to 20:00 GMT+2, featuring a talk titled "Skills are the new Code" by Guy Podjarny, founder and CEO o…
Tessl founder Guy Podjarny, who previously created the developer security platform Snyk, will speak at a fireside chat on 8 April at incident.io's London office at 66 City Rd, discussing how AI is res…
A 2026 data-driven analysis of AI coding tools finds that while 42% of code on GitHub is now AI-generated and 74% of Copilot users report 30-55% faster completion, AI still fails on business logic, se…
Snyk announced the Evo MCP server on September 22, 2026, a remote server that lets developers query AI asset inventory, risk data, policies, and open violations from inside agent harnesses such as Cur…
Unit 42 (Palo Alto Networks) analyzed nearly 50,000 skills from the OpenClaw registry and found that 80% exhibited behavioral deviations, with 18.9% of those deviations traced to clear adversarial int…
A developer published a guide arguing that AI coding agents will silently bypass ten critical software development lifecycle gates—including secret scanning, dependency auditing, type checking, test c…
Snyk released Agent Scan's Skill Inspector, a free tool that scans AI agent skills and MCP servers for malicious code and security vulnerabilities, after analyzing nearly 4,000 agent skills across maj…
Claude 3.5 Sonnet can generate clean code, but shipping it to production without a strict validation pipeline risks subtle regressions and security holes, according to an analysis of agentic coding pr…
Security prevention in agent-generated code is architecturally solved, but applying it inside organizations is difficult because the software development lifecycle is actively shifting, according to a…
A campaign named ClawHavoc uploaded 1,184 malicious skills to ClawHub between late January and early February 2026, with roughly 900 of the platform's 4,500 skills weaponized and five of the seven mos…
Tripwire, an open-source sandboxed security scanner for AI skills and MCP servers, is now available on GitHub, enabling technical teams to discover and scan targets in an isolated Modal sandbox, store…
A new report highlights that advanced AI debugging assistants are significantly improving software development efficiency, with top-tier tools resolving 41% of production bugs within 24 hours compared…
Snyk released Agent Scan, a security scanner that discovers and analyzes AI agent components—including harnesses, MCP servers, and agent skills—for threats such as prompt injections, sensitive data ha…
On August 4, 2026, a malicious commit to the npm package 'keyv' triggered a supply-chain worm that compromised over 400 packages (Elastic) or over 1,300 package versions (Singapore's CSA), totaling mo…
A 2026 guide from nlocoding.com highlights that AI code tools frequently introduce bugs, with 61% of developers reporting weekly issues and 74% of errors going undetected until manual review. The arti…