Vendor
Anthropic vulnerabilities
Updated . Rebuilt every hour.
Junglewise Threat Intelligence has tracked 58 vulnerabilities in Anthropic: 0 in the last 7 days and 13 in the last 90 days, 1 of them critical and 0 exploited in the wild. The most recent, CVE-2026-63128, was published on 16 September 2026. 5 technologies have a page of their own.
- Last 7 days
- 0
- Last 90 days
- 13
- Critical, all time
- 1
- Exploited in the wild
- 0
About Anthropic
AI company that develops large language models and other AI research products.
Anthropic technologies
Latest Anthropic vulnerabilities
- CVE-2026-63128: RMCP is an official Rust SDK for the Model Context Protocol. Prior to 2.0.0, the rmcp crate's stateful Streamable HTTP…highCVSS 7.5EPSS 0.6%
- CVE-2026-63127: RMCP is an official Rust SDK for the Model Context Protocol. Prior to 2.0.0, the rmcp crate's OAuth implementation in…highCVSS 8.2EPSS 0.2%
- CVE-2026-84810: Anthropic claude-skill-antivirus incomplete file analysis in skill scanningmediumCVSS 6.5EPSS 0.3%
- CVE-2026-53965: The MCP PHP SDK (Composer package mcp/sdk) is the official Model Context Protocol SDK for PHP. In versions 0.5.0 through…highCVSS 7.5EPSS 0.6%
- CVE-2026-67432: Model Context Protocol Ruby SDK memory exhaustion in StreamableHTTPTransporthighCVSS 7.5EPSS 0.8%
- CVE-2026-55407: Anthropics Buffa memory exhaustion in protobuf decodermediumCVSS 6.3EPSS 0.8%
- CVE-2026-55406: Anthropics Buffa use-after-free in OwnedViewmediumCVSS 5.9EPSS 0.2%
- CVE-2026-47751: Anthropic Claude Code Action arbitrary code execution in GitHub Actions runnermediumCVSS 5.3
- CVE-2026-59950: modelcontextprotocol python-sdk missing origin validation in WebSocket transporthighCVSS 4EPSS 0.2%
- CVE-2026-52870: Model Context Protocol Python SDK missing authorization in experimental taskshighCVSS 7.6EPSS 0.4%
- CVE-2026-52869: modelcontextprotocol mcp authorization bypass in SSE and HTTP transportshighCVSS 7.1EPSS 0.5%
- CVE-2026-55607: Anthropic Claude Code sandbox escape via git worktree path confusionhighCVSS 4EPSS 0.7%
- CVE-2026-46406: Anthropic Claude Code insecure temporary file in copy commandmediumCVSS 4EPSS 0.1%
- CVE-2026-7574: Anthropic Claude Desktop Cowork missing integrity check in VM image handlinghighCVSS 8.7
- CVE-2026-54316: Anthropic Claude Code data exfiltration via HuggingFace allowlist bypassmediumCVSS 3.1EPSS 0.5%
- CVE-2026-42559: Model Context Protocol RMCP DNS rebinding in Streamable HTTP serverhighCVSS 8.8EPSS 0.2%
- CVE-2026-44470: Anthropic Claude Desktop privilege escalation in CoworkVMServicehighCVSS 7.8EPSS 0.0%
- CVE-2026-44467: Anthropic Claude Desktop SSH host key verification bypassmediumCVSS 6.8EPSS 0.0%
- CVE-2026-40068: Anthropic Claude Code trust bypass via Git worktree spoofinghighCVSS 8.8EPSS 0.6%
- CVE-2026-41686: Anthropic Claude SDK for TypeScript insecure file permissions in LocalFilesystemMemoryToolmediumCVSS 4.4EPSS 0.1%
- CVE-2026-39861: Claude Code: Sandbox Escape via Symlink Following Allows Arbitrary File Write Outside WorkspacecriticalCVSS 10EPSS 0.8%
- CVE-2026-35603: Claude Code: Insecure System-Wide Configuration Loading Enables Local Privilege Escalation on WindowshighCVSS 7.3EPSS 0.1%
- CVE-2026-35022: Anthropic Claude Code OS command injection in authentication helpersinfoCVSS 9.8EPSS 0.6%
- CVE-2026-35021: Anthropic Claude Code OS command injection in prompt editorinfoCVSS 7.8EPSS 0.0%
- CVE-2026-35020: Anthropic Claude Code CLI OS command injection in command lookup helperinfoCVSS 8.4EPSS 0.1%
Most severe Anthropic vulnerabilities
Exploited in the wild first, then by severity and CVSS score.
- CVE-2026-39861: Claude Code: Sandbox Escape via Symlink Following Allows Arbitrary File Write Outside WorkspacecriticalCVSS 10EPSS 0.8%
- CVE-2026-40068: Anthropic Claude Code trust bypass via Git worktree spoofinghighCVSS 8.8EPSS 0.6%
- CVE-2026-42559: Model Context Protocol RMCP DNS rebinding in Streamable HTTP serverhighCVSS 8.8EPSS 0.2%
- CVE-2026-7574: Anthropic Claude Desktop Cowork missing integrity check in VM image handlinghighCVSS 8.7
- CVE-2026-63127: RMCP is an official Rust SDK for the Model Context Protocol. Prior to 2.0.0, the rmcp crate's OAuth implementation in…highCVSS 8.2EPSS 0.2%
- CVE-2026-22561: Anthropic Claude for Windows privilege escalation in installerhighCVSS 7.8EPSS 0.2%
- CVE-2026-44470: Anthropic Claude Desktop privilege escalation in CoworkVMServicehighCVSS 7.8EPSS 0.0%
- CVE-2026-52870: Model Context Protocol Python SDK missing authorization in experimental taskshighCVSS 7.6EPSS 0.4%
- CVE-2026-67432: Model Context Protocol Ruby SDK memory exhaustion in StreamableHTTPTransporthighCVSS 7.5EPSS 0.8%
- CVE-2026-63128: RMCP is an official Rust SDK for the Model Context Protocol. Prior to 2.0.0, the rmcp crate's stateful Streamable HTTP…highCVSS 7.5EPSS 0.6%
Vulnerabilities per week
The last 13 weeks, by the week each vulnerability was published.
| Week of | Bar | Vulns | Critical |
|---|---|---|---|
| 29 Jun 2026 | 2 | 0 | |
| 6 Jul 2026 | 0 | 0 | |
| 13 Jul 2026 | 6 | 0 | |
| 20 Jul 2026 | 0 | 0 | |
| 27 Jul 2026 | 1 | 0 | |
| 3 Aug 2026 | 0 | 0 | |
| 10 Aug 2026 | 0 | 0 | |
| 17 Aug 2026 | 0 | 0 | |
| 24 Aug 2026 | 1 | 0 | |
| 31 Aug 2026 | 1 | 0 | |
| 7 Sep 2026 | 0 | 0 | |
| 14 Sep 2026 | 2 | 0 | |
| 21 Sep 2026 | 0 | 0 |
How this is built
Junglewise Threat Intelligence collects vulnerabilities from NVD, GitHub Security Advisories, OSV, the CISA Known Exploited Vulnerabilities catalog, FIRST EPSS and vendor advisories, and matches each one to the technologies and vendors it affects. Dates are the date a vulnerability was published, in UTC.
The pages are rebuilt from the database every hour. Frozen weekly and monthly reports never change once published, so they can be cited.
Use this data
The same data is at https://junglewise.ai/threats/vendors/anthropic.json, for scripts and language models. It is free to reuse under CC BY 4.0 with a link back to this page.
Cite as: Junglewise Threat Intelligence, "Anthropic vulnerabilities", https://junglewise.ai/threats/vendors/anthropic, 26 September 2026.