Home / Failure Modes / Security Vulnerability

Security Vulnerability

AI Agent Failure Mode — 5 Documented Incidents

What is Security Vulnerability?

Security vulnerability failures occur when an AI coding agent introduces exploitable security flaws such as XSS, SQL injection, hardcoded credentials, authentication bypasses, path traversal, or insecure deserialization. These are high-severity incidents with potential for real-world exploitation.

5
Total Incidents
9.0
Avg Severity /10
4
Agents Affected
3
Critical

Which AI agent security vulnerabilitys the most?

Claude-Code 2 incidents
Github-Copilot 1 incidents
Devin 1 incidents
Unknown-Agent 1 incidents

Frequently Asked Questions

What is security vulnerability in AI coding agents?

Security vulnerability failures occur when an AI coding agent introduces exploitable security flaws such as XSS, SQL injection, hardcoded credentials, authentication bypasses, path traversal, or insecure deserialization. These are high-severity incidents with potential for real-world exploitation.

Which AI agent has the most security vulnerability failures?

Based on 5 documented incidents, Claude-Code has the most security vulnerability failures with 2 incidents, followed by Github-Copilot with 1.

All Security Vulnerability Incidents

STUPID-2026-0004 10.0/10 CRITICAL github-copilot

Copilot autocompleted AWS credentials into public repository

While a developer was writing an AWS configuration file, Copilot suggested a completion that included what appeared to be real AWS access keys. The developer accepted the suggestio...

STUPID-2026-0006 10.0/10 CRITICAL devin

Devin confidently shipped code that passed tests but had a SQL injection vulnerability

Tasked with adding a search feature, Devin built it using string concatenation for SQL queries instead of parameterized queries. All functional tests passed because the tests didn'...

STUPID-2026-0022 10.0/10 CRITICAL unknown-agent

AI vibe-coded Next.js app pinned vulnerable dependency — cryptominer compromised production server

A developer used an AI coding agent to build a Next.js web service via 'vibe coding' (building from functional descriptions). The agent pinned a dependency version that contained C...

STUPID-2026-0019 7.5/10 HIGH claude-code

Claude Opus 4.5 leaked API key in console logs during YouTube scraper build

While building a YouTube scraper, Claude Opus 4.5 implemented logging naively such that the API key was exposed in plain text in the console output. The developer had to add explic...

STUPID-2026-0024 7.5/10 HIGH claude-code

Claude Code MCP trust boundary failures allow workspace privilege escalation

Security researcher Jashid Sany documented three systemic trust boundary failures in Claude Code v2.1.63 related to the Model Context Protocol (MCP): (1) weak MCP server configurat...