ai-coding-assistants
Safeguard articles tagged "ai-coding-assistants" — guides, analysis, and best practices for software supply chain and application security.
13 articles
Every AI Coding Tool Has the Same Vulnerability, and It Isn't a Bug
Sandbox escapes in Claude Code, critical CVEs in Cursor, a 10.0 in Gemini CLI, prompt injection in Copilot. Different vendors, one shared cause: the agent must hold elevated access to be useful.
Security risks introduced by AI coding assistants and gen...
AI coding assistants now write huge shares of production code. Real 2025 incidents show hallucinated packages, leaked secrets, and vulnerable defaults ship with it.
The Jscrambler npm Compromise Went After Your AI Coding Assistant's Credentials
On 11 July 2026, five versions of the jscrambler package plus its webpack, gulp, grunt and metro plugins shipped malicious native binaries. The payload targeted crypto wallets and the credential stores of Claude Desktop, Cursor and Windsurf — and later versions fired on import, not install, defeating --ignore-scripts.
The supply-chain and IP risk hiding inside AI coding assistants
GitHub has disclosed that Copilot suggestions match training-set code verbatim about 1% of the time — and a class action over it is still being argued in 2026.
The Hidden Risks of AI Coding Assistants
A 2021 NYU study found 40% of Copilot-generated code contained exploitable bugs — and that's before counting leaked secrets or hallucinated packages.
Securing AI Coding Assistants: Guardrails That Hold
AI coding assistants are in nearly every IDE now. Banning them fails; trusting them blindly fails harder. The middle path is guardrails — technical controls that let assistants move fast without letting them ship the wrong thing.
Autocomplete Anxiety: Measuring How Often AI Coding Assis...
Studies show 40-45% of AI-suggested code contains exploitable flaws, and models hallucinate fake packages developers install. Here's what the data says.
The Prompt Injection Problem Hiding Inside Everyday Code ...
AI coding assistants read untrusted files as instructions, not data. Here's how prompt injection sneaks malicious code into your commits — and how to catch it before it ships.
Comparing Insecure Output Rates Across Popular AI Coding ...
A benchmark-driven look at insecure output rates across GitHub Copilot, Cursor, Amazon Q, and Tabnine, and why the model matters more than the brand.
TrapDoor: The Cross-Ecosystem Crypto Stealer That Targeted DeFi Developers (May 2026)
Socket disclosed TrapDoor on May 24, 2026: 34+ malicious packages and 384+ versions across npm, PyPI, and Crates.io built to steal crypto wallets, SSH keys, and cloud credentials from crypto, DeFi, Solana, and AI developers.
Are AI Coding Assistant Vendors Ready to Own Their Securi...
AI coding assistants ship indemnification for copyright suits, not for the vulnerabilities they introduce. Here's the liability gap enterprises need to understand.
Supply Chain Risks of AI Coding Assistants
Copilot, Cursor, and Claude Code change what enters your codebase and how. A practitioner's map of the real supply chain risks — hallucinated packages, rules-file injection, and unreviewed transitive trust.
Self-healing security runs on Safeguard.
Your first fix PR is minutes away.
No sales call required, even your agent can complete the purchase over MCP.