CVE-2026-96620

criticalpriority · track · 0EPSS 0.007 · 53 %ileUS
grimoire priorityP2 · Scheduled
Exploitable?
—
no known exploitation
Likely / severe?
0.70% (53rd) EPSS
CVSS 9.2 · critical
Is it mine?
1 products
Check your exposure
Who says so?
1 sources
osv

EPSS · scored 2026-10-27

Claude Agent SDK is a Python SDK for Claude Code. Prior to 0.2.121, the resume option is passed to the Claude CLI as a separate token without validation, so a dash-leading value is parsed as an independent CLI flag. An attacker can inject the --mcp-config flag to define a stdio MCP server, causing arbitrary command execution at CLI startup outside the SDK permission controls. The issue is exploitable when an application passes untrusted input as the session ID to resume. This issue is fixed in version 0.2.121.

01

cvss

Technical severity, 0–10. One row per scoring source.

9.2
010

base score 0–10 · higher = more severe · hover a vector to decode⚠ sources disagree · high–critical · 7.2–9.2

v4.09.2criticalCVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:Xsecondary · security-advisories@github.com
v4.07.2highCVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:U/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:Xsecondary · CVSS-BT (t0sche)
02

ssvc

A stakeholder decision — act, attend, or track — from exploitation and impact.

Exploitation: none · Automatable: yes · Technical Impact: totalcoordinatorcisa_ssvcUS
03

weaknesses

The CWE weakness classes this flaw belongs to.

CWE-20USCWE-20USCWE-20USCWE-78USCWE-78USCWE-78USCWE-88USCWE-88USCWE-88US
04

affected products (1)

Products and version ranges known to be affected.

05

advisories (1)

Vendor and ecosystem advisories describing this.

  • GHSA-h4mw-j7qp-8mwmClaude Agent SDK: Argument Injection via resume Option Allows Arbitrary Command Execution2026-10-16ghsa · osvUS

    The Claude Agent SDK for Python passed the `resume` option to the Claude CLI without validation, allowing a value beginning with `-` to be interpreted as a separate CLI flag. An attacker who controlled the `resume` value could inject an `--mcp-config` flag defining a stdio MCP server, resulting in arbitrary command execution at CLI startup that bypassed the SDK's permission controls. Exploitation required an application to pass untrusted input as the session ID to resume. Users are advised to update to the latest version of the claude-agent-sdk package. Thank you to hackerone.com/moamenmahmood for reporting this issue.

06

sources

Which sources assert this record, and when.

osvNVDCVE.orgMITRE
published 2026-10-16modified 2026-10-21
07

references (4) nvd

External citations — patches, advisories, analysis.

scry

Attackers weaponise vulnerabilities like this against internet-exposed services. See what is exposed on your perimeter with a free passive scan.

Free scan