CVE-2026-9367
HighCVSS 7.3Exploitation Probability (EPSS)
Elevated risk74th percentile - higher than 74% of all known CVEs
Summary
A command injection vulnerability was found in NousResearch hermes-agent up to commit 5157f5427f19488b31c6fdebbacd15d798ce7f63 in the detect_dangerous_command function of tools/approval.py. The attack can be performed remotely. The vendor did not respond.
Risk Assessment
A remote attacker can execute arbitrary system commands, potentially leading to full system compromise.
Recommendation
Update hermes-agent to a patched version or apply other mitigations, as the vendor has not released a fix.
Other vulnerabilities in NousResearch hermes-agent
See all- CVE-2026-18773Medium
A vulnerability was detected in NousResearch hermes-agent up to 2026.6.5. The issue affects the _check_slash_access function in gateway/run.py of the Quick Command Handler component. Manipulation results in incorrect authorization. The attack can be launched remotely. The exploit is public and may be used. The vendor was contacted but did not respond.
- CVE-2026-15311Low
A Cross-Site Scripting (XSS) vulnerability was found in the Matrix Adapter component of NousResearch hermes-agent up to version 2026.5.29.2, specifically in the MatrixAdapter._markdown_to_html function of gateway/platforms/matrix.py. The attack can be executed remotely and the exploit is publicly available.
- CVE-2026-14627Medium
A security vulnerability was found in NousResearch hermes-agent up to version 0.15.2. The issue affects the DiscordAdapter._is_allowed_user function in the gateway/platforms/discord.py file, leading to improper authentication. The attack can be launched remotely but is difficult to exploit.
- CVE-2026-14625Medium
A security vulnerability has been found in NousResearch hermes-agent up to version 0.15.2 in the shell.exec function of tui_gateway/server.py. It allows remote bypass of protection mechanisms. The exploit is public and the vendor did not respond.
- CVE-2026-14617Low
A security vulnerability has been detected in the Streaming Reasoning Tag Filter component of NousResearch hermes-agent up to version 2026.4.30, involving improper handling of case sensitivity in the GatewayStreamConsumer._filter_and_accumulate function. The attack can be initiated remotely but is difficult to exploit due to high complexity. The project decided not to implement a dedicated fix, citing maintenance costs.
- CVE-2026-10224Medium
A vulnerability was detected in NousResearch hermes-agent up to version 2026.4.30. It affects the _handle_webhook_request function in the file gateway/platforms/feishu.py of the Webhook Endpoint component. Manipulation leads to resource consumption. The attack can be launched remotely. The exploit has been publicly disclosed.
- CVE-2026-9368High
A vulnerability in NousResearch hermes-agent up to version 2026.4.16 affects the execute_code function in the file tools/code_execution_tool.py of the Environment Variable Handler component. Manipulation leads to a sandbox issue. The attack can be launched remotely, and the exploit is publicly available.
- CVE-2026-9366High
A vulnerability was found in NousResearch hermes-agent version 2026.4.23 in the function _scan_context_content of the file agent/prompt_builder.py. The manipulation results in injection. The attack may be performed remotely. The exploit has been made public and could be used. The vendor was contacted but did not respond.
- CVE-2026-9353High
A vulnerability was found in NousResearch hermes-agent up to version 2026.4.23 in an unknown function of the file agent/skills_guard.py in the Skills Guard Multi-Word Prompt Handler component. Manipulation of the argument THREAT_PATTERNS leads to injection. Remote exploitation is possible. The exploit has been disclosed publicly and may be used. The vendor was contacted but did not respond.
- CVE-2026-9352Medium
A weakness has been identified in NousResearch hermes-agent up to version 2026.4.23 in the _make_run_env function of the tools/environments/local.py file in the Messaging Gateway Handler component. Manipulation can lead to information disclosure. The attack can be launched remotely, and the exploit has been made public and could be used for attacks. The vendor did not respond.
Original NVD description (English source)
A vulnerability was determined in NousResearch hermes-agent up to 5157f5427f19488b31c6fdebbacd15d798ce7f63. This affects the function detect_dangerous_command of the file tools/approval.py of the component terminal_tool. This manipulation causes os command injection. It is possible to initiate the attack remotely. The exploit has been publicly disclosed and may be utilized. The vendor was contacted early about this disclosure but did not respond in any way.

