{
 "license": "CC-BY-4.0",
 "attribution": "Fide AI, Agentic Cyber Explorer",
 "url": "https://agentic-cyber-explorer.pages.dev/events/anthropic-claude-code-security-2026/",
 "asOf": "2026-09-26",
 "id": "anthropic-claude-code-security-2026",
 "date": "2026-02-20",
 "datePrecision": "day",
 "title": "Anthropic releases Claude Code Security in limited preview to scan code and propose patches",
 "lane": "defense",
 "kind": "tool-release",
 "summary": "Anthropic released Claude Code Security as a limited research preview for Enterprise and Team customers, with expedited free access for open-source maintainers. The tool reasons about data flow across a codebase, re-examines each finding in a multi-stage verification pass, assigns severity and confidence ratings, and proposes patches that are applied only with human approval.",
 "whyItMatters": "It packages frontier-model vulnerability finding for defenders with explicit human approval gates, amid concerns about the same capability aiding attackers.",
 "actors": [
  "anthropic"
 ],
 "topics": [
  "vulnerability-discovery",
  "vulnerability-repair"
 ],
 "atlas": [
  "human-approver"
 ],
 "artifacts": [
  "claude-code-security",
  "claude-opus-4"
 ],
 "sources": [
  {
   "url": "https://www.anthropic.com/news/claude-code-security",
   "publisher": "Anthropic",
   "title": "Making frontier cybersecurity capabilities available to defenders",
   "date": "2026-02-20",
   "type": "primary",
   "accessed": "2026-09-25"
  }
 ],
 "keyFacts": [
  {
   "fact": "Released as a limited research preview to Enterprise and Team customers, with expedited access for open-source maintainers.",
   "locator": "Opening and 'Getting started'"
  },
  {
   "fact": "Every finding goes through multi-stage verification where Claude tries to prove or disprove it; nothing is applied without human approval.",
   "locator": "Section 'How Claude Code Security works'"
  },
  {
   "fact": "Anthropic restates that Opus 4.6 found over 500 vulnerabilities in production open-source codebases.",
   "locator": "Section 'Using Claude for cybersecurity'"
  }
 ],
 "significance": 3,
 "fideQuestions": [
  "FID-076",
  "FID-088"
 ],
 "methods": [
  "ai-vulnerability-discovery",
  "automated-patching",
  "human-approval",
  "patch-verification"
 ],
 "review": "assistant-drafted",
 "addedOn": "2026-09-25"
}