Goal Reached Thanks to every supporter — we hit 100%!

Goal: 1000 CNY · Raised: 1359 CNY

100%
Buy Us a Coffee

CVE-2026-9353— NousResearch hermes-agent Skills Guard Multi-Word Prompt skills_guard.py injection

Quick assessment

Affected
NousResearch hermes-agent
Exploitation
No confirmed in-the-wild exploitation; assess based on exposure
Recommended action
Check the vendor advisory and references for a fixed version. If immediate upgrade is impossible, restrict exposure and increase monitoring.

Hermes Agent是Nous Research开源的一款具备自我学习循环的AI代理工具。 Hermes Agent 2026.4.23及之前版本存在安全漏洞，该漏洞源于Skills Guard Multi-Word Prompt Handler组件文件agent/skills_guard.py中未知函数对参数THREAT_PATTERNS的操作，可能导致注入。

CVSS 7.3 · High EPSS 0.52% · P43

Possible ATT&CK Techniques 1 AI

T1190 · Exploit Public-Facing Application

Affected Version Matrix 24

VendorProduct Version RangeStatus
NousResearch hermes-agent 2026.4.0 affected
2026.4.1 affected
2026.4.2 affected
2026.4.3 affected
2026.4.4 affected
2026.4.5 affected
2026.4.6 affected
2026.4.7 affected
… +16 more rows
Get alerts for future matching vulnerabilities Log in to subscribe

I. Basic Information for CVE-2026-9353

Vulnerability Information
Have questions about the vulnerability? See if Shenlong's analysis helps!
View Shenlong Deep Dive ↗

Although we use advanced large model technology, its output may still contain inaccurate or outdated information.Shenlong tries to ensure data accuracy, but please verify and judge based on the actual situation.

Vulnerability Title
NousResearch hermes-agent Skills Guard Multi-Word Prompt skills_guard.py injection
Source: CVE Program / CVE List V5
Vulnerability Description
A security vulnerability has been detected in NousResearch hermes-agent up to 2026.4.23. Impacted is an unknown function of the file agent/skills_guard.py of the component Skills Guard Multi-Word Prompt Handler. The manipulation of the argument THREAT_PATTERNS leads to injection. Remote exploitation of the attack is possible. The exploit has been disclosed publicly and may be used. The vendor was contacted early about this disclosure but did not respond in any way.
Source: CVE Program / CVE List V5
CVSS Information
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:L
Source: CVE Program / CVE List V5
Vulnerability Type
输出中的特殊元素转义处理不恰当(注入)
Source: CVE Program / CVE List V5
Vulnerability Title
Hermes Agent 安全漏洞
Source: CNNVD (China National Vulnerability Database)
Vulnerability Description
Hermes Agent是Nous Research开源的一款具备自我学习循环的AI代理工具。 Hermes Agent 2026.4.23及之前版本存在安全漏洞，该漏洞源于Skills Guard Multi-Word Prompt Handler组件文件agent/skills_guard.py中未知函数对参数THREAT_PATTERNS的操作，可能导致注入。
Source: CNNVD (China National Vulnerability Database)
CVSS Information
N/A
Source: CNNVD (China National Vulnerability Database)
Vulnerability Type
N/A
Source: CNNVD (China National Vulnerability Database)

Shenlong Deep Dive — AI Deep Analysis

10-question deep dive: root cause, exploitation, mitigation, urgency. Read summary free, full version requires login.

Read summary Full analysis (login)

Affected Products

Vendor Product Affected Versions CPE Subscribe
NousResearch hermes-agent 2026.4.0 cpe:2.3:a:nousresearch:hermes-agent:*:*:*:*:*:*:*:*

II. Public POCs for CVE-2026-9353

# POC Description Source Link Shenlong Link
AI-Generated POC Premium

No public POC found.

Login to generate AI POC

III. Intelligence Information for CVE-2026-9353

请登录查看更多情报信息。

Proof of Concept for CVE-2026-9353 (1)

News Coverage for CVE-2026-9353 (1)

Same Patch Batch · NousResearch · 2026-05-24 · 9 CVEs total

CVE-2026-9350 7.3 HIGH NousResearch hermes-agent Batch Runner approval.py check_all_command_guards authorization
CVE-2026-9366 7.3 HIGH NousResearch hermes-agent prompt_builder.py _scan_context_content injection
CVE-2026-9368 7.3 HIGH NousResearch hermes-agent Environment Variable code_execution_tool.py execute_code sandbox
CVE-2026-9367 7.3 HIGH NousResearch hermes-agent terminal_tool approval.py detect_dangerous_command os command in
CVE-2026-9351 6.5 MEDIUM NousResearch hermes-agent read_file Tool file_tools.py _is_blocked_device path traversal
CVE-2026-9354 6.5 MEDIUM NousResearch hermes-agent Slack Agent/Mattermost Agent escape output
CVE-2026-9352 5.3 MEDIUM NousResearch hermes-agent Messaging Gateway local.py _make_run_env information disclosure
CVE-2026-9369 5.3 MEDIUM NousResearch hermes-agent CLI web-dashboard web_server.py _discover_dashboard_plugins comp

IV. Related Vulnerabilities

Same product: hermes-agent
Same vendor: NousResearch
Same weakness: CWE-74

V. Comments for CVE-2026-9353

No comments yet

Leave a comment