{
  "type": "post",
  "slug": "claude-mythos-5-comes-to-claude-security",
  "title": "Claude Mythos 5 comes to Claude Security",
  "description": "Anthropic moved Mythos-class cyber capability into Claude Security behind safety classifiers, turning a model's offensive potential into an access-control decision rather than a release note.",
  "retrieval_nugget": "Anthropic moved Mythos-class cyber capability into Claude Security behind safety classifiers, turning a model's offensive potential into an access-control decision rather than a release note.",
  "published_at": "2026-08-29",
  "updated_at": "2026-09-12",
  "record_date": "2026-08-21",
  "date_kind": "published_at",
  "topics": [
    "agent-security",
    "agents",
    "ai",
    "security"
  ],
  "entities": [
    "Claude"
  ],
  "source_url": "https://claude.com/blog/bringing-claude-mythos-5-to-more-defenders",
  "source_title": "Bringing the cybersecurity capabilities of Claude Mythos 5 to more defenders",
  "source_domain": "claude.com",
  "source_terms": [
    "Claude",
    "Mythos",
    "Security",
    "defenders",
    "cyber"
  ],
  "summary_word_count": 197,
  "schema_version": "newruntime-agent-readable-v0.2",
  "stable_id": "post:claude-mythos-5-comes-to-claude-security",
  "status": "published",
  "source_urls": [
    "https://claude.com/blog/bringing-claude-mythos-5-to-more-defenders"
  ],
  "visuals": [],
  "editorial_provenance": {
    "schema_version": "newruntime-editorial-copy-v1",
    "content_status": "source_grounded_final",
    "final_copy_sha256": "sha256:73cf422a1151b514622ca80196e2e50bada60cd7cc6ffc60a01c1511475b1f76",
    "reviewed_at": "2026-09-12T10:00:00Z",
    "source_evidence_count": 1,
    "verified_claim_count": 2,
    "site_analysis_schema_version": "newruntime-site-analysis-v1",
    "site_object_kind": "field_note",
    "observed_fact_count": 2,
    "implication_count": 1,
    "watch_condition_count": 1,
    "related_record_count": 2
  },
  "analysis": {
    "schema_version": "newruntime-site-analysis-v1",
    "object_kind": "field_note",
    "thesis": "Anthropic moved Claude Mythos 5 into Claude Security on 21 August and said it will reach partners' cyber defence tools soon, alongside a $35M fund for open-source software security and an expansion of its Cyber Verification Program.",
    "observed_facts": [
      {
        "text": "Claude Mythos 5 is now available in Claude Security and is described as coming soon to partners' cyber defence tools.",
        "source_urls": [
          "https://claude.com/blog/bringing-claude-mythos-5-to-more-defenders"
        ]
      },
      {
        "text": "Project Glasswing launched in April to put Mythos-class models with a small group of organisations securing critical software.",
        "source_urls": [
          "https://claude.com/blog/bringing-claude-mythos-5-to-more-defenders"
        ]
      }
    ],
    "mechanism": "Anthropic describes safety classifiers and safeguards that let Mythos-class access widen without handing over offensive cyber work, with Claude Fable 5 as the first step: broadly available while dual-use cyber work stayed blocked.",
    "why_now": "since April, Project Glasswing had confined Mythos-class capability to a small group of organisations defending critical software, deliberately giving defenders a window to find and fix vulnerabilities before comparable capability became generally available or reached malicious actors",
    "implications": [
      "The staged rollout treats a model's offensive potential as an access-control problem to be managed inside product surfaces, not as a property to be published and hoped about."
    ],
    "evidence_boundary": "This is a vendor announcement: it states availability, a fund, and programme intent, not measured defensive outcomes.",
    "watch_conditions": [
      "The condition to watch is whether the Cyber Verification Program publishes what a verified defender must satisfy, because access gated by a classifier is only auditable when the entry criteria are visible."
    ],
    "related_records": [
      {
        "url": "https://newruntime.com/posts/vm-isolation-is-only-one-agent-containment-boundary",
        "relation": "Prior coverage of containment boundaries for cyber-capable agents."
      },
      {
        "url": "https://newruntime.com/topics/agent-security",
        "relation": "Topic hub collecting agent access and authority control points."
      }
    ]
  },
  "routes": {
    "html": "https://newruntime.com/posts/claude-mythos-5-comes-to-claude-security/",
    "markdown": "https://newruntime.com/posts/claude-mythos-5-comes-to-claude-security.md",
    "json": "https://newruntime.com/posts/claude-mythos-5-comes-to-claude-security.json"
  }
}