Vulnerability · Published May 24, 2026 · Assigned by VulDB

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

High 7.3

In plain English

A high-severity injection in hermes-agent from NousResearch. Proof-of-concept exploit code is public. The chance it is exploited in the next 30 days is estimated at 0.3% (higher than 23% of all known vulnerabilities). Fixed versions are available - upgrade the affected package.

Severity (CVSS 3.1)7.3High

Exploit odds (30 days)0.3%top 76.6% of all CVEs

Exploited in the wildNo reportsCISA SSVC: poc

Attack toolingNoneno public tooling known

Official 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.

Affected packages & fixed versions

Ecosystem	Package	Fixed in	Advisory
PyPI	hermes-agent	0.15.0	GHSA-238w-f66p-w349
PyPI	hermes-agent	0.15.0	PYSEC-2026-2509

Affected products (as reported)

Vendor	Product	Versions
NousResearch	hermes-agent	2026.4.02026.4.12026.4.22026.4.32026.4.42026.4.52026.4.62026.4.7

Weakness: CWE-74, CWE-707 · Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:L/E:P/RL:X/RC:R

References

VDB-365316 | NousResearch hermes-agent Skills Guard Multi-Word Prompt skills_guard.py injection vdb-entry, technical-description
VDB-365316 | CTI Indicators (IOB, IOC, TTP, IOA) signature, permissions-required
Submit #812216 | NousResearch hermes-agent 2026.4.23 Injection (CWE-74) third-party-advisory
https://gist.github.com/YLChen-007/82a3539d6358842e69dfaef0a9fcf14a exploit