---
title: "CVE-2026-96620: Claude Agent SDK: Argument Injection via resume Option Allows Arbitrary Command Execution"
url: https://erroneo.us/security/cve/CVE-2026-96620/
description: "CVE-2026-96620 (claude-agent-sdk (PyPI)): Claude Agent SDK is a Python SDK for Claude Code. Prior to 0.2.121, the resume option is passed to the Claude CLI…"
type: vulnerability-record
section: "Security Digest"
date_published: 2026-10-16
date_modified: 2026-10-21
publisher: erroneo.us
author: erroneo.us Security Desk
cve: "CVE-2026-96620"
kev: false
cvss: 9.2
epss: 0.00578
license: All rights reserved. Quote with attribution and a link.
---

# CVE-2026-96620: Claude Agent SDK: Argument Injection via resume Option Allows Arbitrary Command Execution

**Signal:** Critical and remotely reachable. No known exploitation yet, but it is network-reachable without credentials. (Computed from CISA KEV, EPSS and CVSS; not an editorial verdict.)

## Should you act?

- No confirmed exploitation. It is not in CISA’s KEV catalog and CISA’s SSVC assessment records no known exploitation. That can change quickly after a public write-up; re-check if you are affected.
- EPSS 0.6% , percentile 47: FIRST’s model estimates a 0.6% chance of exploitation activity in the next 30 days, higher than 47% of all scored CVEs (low). EPSS is a probability, not a verdict.
- Remotely reachable with no login and no user interaction. If the vulnerable service is exposed to the internet, prioritize it.
- Only applies if your dependency tree includes claude-agent-sdk (PyPI) < 0.2.121, fixed in 0.2.121 . Check your lockfile, including transitive dependencies.

## What it is

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.

## Affected packages

- PyPI `claude-agent-sdk` < 0.2.121 → fixed: 0.2.121

## Affected products (CVE record)

- anthropics claude-agent-sdk-python: < 0.2.121

## Severity

- GitHub_M CVSS 4.0: 9.2 `CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N`
- GitHub CVSS 4.0: 9.2 `CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N`

## Exploitation

- CISA KEV: not listed (no confirmed in-the-wild exploitation known to CISA).
- CISA SSVC assessment: exploitation none , automatable yes , technical impact total .
- FIRST EPSS (model of 2026-10-21): 0.58% probability of exploitation activity in the next 30 days; percentile 47.

## Sources

- [GitHub Advisory GHSA-h4mw-j7qp-8mwm](https://github.com/advisories/GHSA-h4mw-j7qp-8mwm)
- [CVE.org record CVE-2026-96620](https://www.cve.org/CVERecord?id=CVE-2026-96620)
- [NVD entry CVE-2026-96620](https://nvd.nist.gov/vuln/detail/CVE-2026-96620)
- [github.com](https://github.com/anthropics/claude-agent-sdk-python/security/advisories/GHSA-h4mw-j7qp-8mwm)
- [github.com](https://github.com/anthropics/claude-agent-sdk-python/pull/1123)
- [github.com](https://github.com/anthropics/claude-agent-sdk-python/commit/347a1cb88d7652199399db1b7f914e0a459a6662)
- [github.com](https://github.com/anthropics/claude-agent-sdk-python/releases/tag/v0.2.121)

---
Source page: https://erroneo.us/security/cve/CVE-2026-96620/ · Security Digest: https://erroneo.us/security/ · More: https://erroneo.us/llms.txt