Skip to main content

Hermes Agent EUVDEUVD-2026-67867

| CVE-2026-82021 CRITICAL
Download of Code Without Integrity Check (CWE-494)
2026-08-28 VulnCheck GHSA-w9gw-vgmg-q67h
9.0
CVSS 4.0 · Vendor: VulnCheck
View

Attack ChainAIDerived

Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain

  1. Access
    Compromise upstream repo write access
  2. Delivery
    Commit malicious code to tracked branch
  3. Exploit
    Operator installs/updates catalog entry
  4. Execution
    Host pulls unpinned branch code
  5. Persist
    Execute arbitrary code on host
  6. Impact
    Auto-propagate to all installers

Vulnerability Assessment

Weakness
CWE
OWASP Top 10:2025
A08:2025 - Software or Data Integrity Failures
ASVS 5.0
V15 - Secure Coding and Architecture Code integrity verification
Exploitation Conditions

Exploitation requires (1) the attacker to first compromise the specific third-party upstream repository that a Hermes Agent MCP catalog entry references, and (2) that entry to be resolved via a mutable branch reference rather than a pinned commit SHA - the exact insecure configuration named in the description. A victim operator must install or update the affected catalog entry (the passive user interaction reflected in UI:P), after which propagation is automatic with no further operator action. Limiting factors: only Hermes Agent 0.18.2-0.18.x (before 0.19.0) are affected, the upstream-compromise prerequisite makes this high-complexity (AC:H) rather than trivially exploitable, and hosts that never install the vulnerable catalog entry are not exposed.

Impact

Signals are mixed and warrant careful reading. The vendor CVSS 4.0 base score is 9.0 (Critical) with AV:N and full High impact to both the vulnerable and subsequent systems (VC/VI/VA and SC/SI/SA all High), reflecting the fan-out of a supply-chain compromise. However AC:H and UI:P materially constrain real-world exploitation: the attacker must first compromise the upstream repository (non-trivial) and a victim must install/update the catalog entry. There is no EPSS score in the provided data, no KEV listing, and no POC reference, so this reads as a high-impact-but-precondition-heavy issue rather than a demonstrated, widely-exploited one. Note a tension worth flagging: the description states 'no further action required by the operator' (implying autonomous propagation) while the CVSS vector carries UI:P - reconcile by treating the initial catalog install/update as the passive user interaction, after which propagation is automatic. Verify EPSS/POC status with VulnCheck before final prioritization.

Exploit ScenarioDerived

An attacker gains write access to a third-party upstream repository that a Hermes Agent MCP catalog entry tracks by branch, then commits malicious code to that branch head. When operators install or update the affected catalog entry, their hosts pull and execute the attacker's code, yielding remote code execution across every installing host. No public POC is identified at time of analysis, and the attack's difficulty is dominated by the upstream-compromise step (CVSS AC:H).

Act Now Prioritize within 24-48h. Why: CRITICAL severity.
Patch available 0.2% exploitation probability (EPSS)

Get an e-mail when this CVE is exploited, patched or re-scored. Free account.

Alert me

AnalysisAIv1 – Aug 28, 2026

Supply-chain remote code execution affects NousResearch Hermes Agent versions 0.18.2 through 0.18.x (prior to 0.19.0), where the bundled MCP catalog references a third-party upstream repository via a mutable git branch instead of a pinned commit SHA. An attacker who compromises that upstream repository can push malicious code that is automatically pulled onto every host installing the affected catalog entry, achieving code execution with no further operator action. Reported by VulnCheck; a vendor patch exists (v0.19.0) but there is no public exploit identified at time of analysis and it is not listed in CISA KEV.

RemediationAI

Vendor-released patch: upgrade to Hermes Agent v0.19.0 (release tag v2026.7.20), which is available from the vendor and contains the fix (commit 9df5f879b4a5925c0f8f947e7e16ed8e845932c3, PR #64463) - see https://github.com/NousResearch/hermes-agent/releases/tag/v2026.7.20. If you cannot upgrade immediately, remove or disable the affected MCP catalog entry that resolves an upstream repository via a mutable branch, or pin the catalog entry to a specific trusted commit SHA rather than a branch so unreviewed upstream changes cannot be pulled; the trade-off is that pinned entries no longer receive automatic upstream updates and must be manually reviewed and bumped. Additionally restrict which catalog sources hosts may install from and monitor for unexpected catalog updates. Full details in the VulnCheck advisory: https://www.vulncheck.com/advisories/hermes-agent-mcp-catalog-supply-chain-rce-via-mutable-branch-reference.

Affected VersionsENISA

hermes-agent 2026.7.7.2 <2026.7.20
hermes-agent 0.18.2 <0.19.0

Affected Packages

euvd hermes_agent fix: 0.19.0, 2026.7.20

Source: OSV (Open Source Vulnerabilities)

Affects your stack

Affects 1 package in euvd

Checking this CVE against your own stacks and servers is part of the Basic plan and up.

Technical ContextAI · show details

The affected component is the Model Context Protocol (MCP) catalog shipped with Hermes Agent, an AI agent runtime by NousResearch (CPE cpe:2.3:a:nousresearch:hermes-agent). The catalog resolves upstream tool/server definitions by tracking a mutable branch reference (a moving branch tip) rather than an immutable commit SHA, so the content pulled at install time is whatever currently sits at the branch head. This maps directly to CWE-494 (Download of Code Without Integrity Check): code is fetched and executed without verifying it against a known-good, pinned integrity anchor. Because the trust boundary extends to an external repository the operator does not control, a compromise upstream is inherited by all downstream installs.

Know when Hermes Agent gets its next vulnerability

  • Watch products and get an email when something new is published
  • Check this CVE against your own stack and servers
  • Exploit and PoC references, threat intelligence from 22 sources

No credit card · 7-day full trial

Severity by source

Vendor (VulnCheck) PRIMARY 9.0 CRITICAL
CVSS:4.0/AV:N/AC:H/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H
vuln.today AI 8.3 HIGH

Attacker needs no victim-side privileges (PR:N) but must compromise upstream (AC:H) and the operator must install the entry (UI:R); the external-repo trust boundary crossing gives S:C with full RCE impact.

3.1 AV:N/AC:H/PR:N/UI:R/S:C/C:H/I:H/A:H
4.0 AV:N/AC:H/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H

Primary rating from Vendor (VulnCheck).

CVSS VectorVendor: VulnCheck

Attack Vector
Network
Attack Complexity
High
Attack Requirements
None
Privileges Required
None
User Interaction
Passive
Vulnerable system: Confidentiality
High
Vulnerable system: Integrity
High
Vulnerable system: Availability
High
Subsequent system: Confidentiality
High
Subsequent system: Integrity
High
Subsequent system: Availability
High

DescriptionCVE.org

Hermes Agent 0.18.2 prior to 0.19.0 contains a supply chain vulnerability in its bundled MCP catalog that allows a remote attacker to execute arbitrary code by compromising a third-party upstream repository referenced via a mutable branch rather than a pinned commit SHA. An attacker who compromises the upstream repository can propagate malicious code to every host that installs the affected catalog entry, with no further action required by the operator.

Executive ViewAI

Why It Matters

NousResearch Hermes Agent versions 0.18.2 through 0.18.x contain a supply-chain remote code execution vulnerability affecting any organization deploying the affected software catalog. The vulnerability allows an attacker who compromises the upstream dependency repository to automatically push and execute malicious code on every host running these versions, without requiring user interaction or authentication, creating a pervasive infrastructure compromise risk.

Business Impact: Severe
Compensating Controls

Apply vendor patch.

Threat Intelligence Overview

EPSS Score
0.2%
Top 87% exploitable
Prevalence
Low
niche AI agent runtime from a single vendor, limited enterprise footprint
Data Quality
7 sources
Confidence: High

How this decision was reached

The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.

SignalValue
Severity CRITICAL
CISA KEV No
Public POC No
EPSS 0.2%
Patch available Yes
Active threat No
Deciding rule
CRITICAL severity
What every action level means
Emergency Immediate action required. Criteria: CRITICAL + in CISA KEV (actively exploited) without patch, OR CRITICAL + EPSS > 30% without patch, OR CRITICAL + active threat signals without patch.
Act Now Prioritize within 24-48h. Criteria: in KEV with patch available, OR CRITICAL + public POC, OR HIGH + EPSS > 10%, OR CRITICAL + active threat, OR any CRITICAL severity.
This Week Schedule remediation this week. Criteria: HIGH severity (CVSS 7.0-8.9). Patch within normal change window.
This Month Plan within 30 days. Criteria: MEDIUM severity (CVSS 4.0-6.9), lower exploitation probability. Include in regular patch cycle.
Monitor Track and reassess. Criteria: LOW severity (CVSS 0.1-3.9), minimal exploitation risk. No immediate action needed.

SSVC AssessmentENISA

SSVC Decision
Scheduled
Exploitation
None
System Exposure· derived
Open
Automatable
No
Human Impact
Very High
Technical Impact
Total

Priority Score

45
Low Medium High Critical
KEV: 0
EPSS: 0
CVSS: +45
PoC: 0
Weaponized: 0
Threat intel: 0
Gov advisory: 0

Attack Technique

Remote Code Execution represents the critical moment when an attacker successfully runs arbitrary code on a target system without physical access.

CVE-2026-71963 HIGH POC
8.6 Sep 03

Remote code execution in NousResearch Hermes Agent 0.18.2-0.21.0 allows an unauthenticated attacker to execute arbitrary

CVE-2026-9367 MEDIUM POC
5.5 May 24

Remote command injection in NousResearch hermes-agent allows unauthenticated attackers to execute arbitrary OS commands

CVE-2026-14628 MEDIUM POC
5.5 Jul 04

Path traversal in NousResearch hermes-agent (all versions through 2026.5.16) exposes arbitrary server-side files via the

CVE-2026-9351 MEDIUM POC
5.5 May 24

Path traversal in NousResearch hermes-agent through version 2026.4.16 allows remote unauthenticated attackers to bypass

CVE-2026-9368 MEDIUM POC
5.5 May 24

Remote sandbox escape in NousResearch hermes-agent versions up to 2026.4.16 allows unauthenticated attackers to manipula

CVE-2026-7396 MEDIUM POC
5.5 Apr 29

Path traversal in NousResearch hermes-agent 0.8.0 exposes arbitrary server-side files to unauthenticated remote attacker

CVE-2026-10221 MEDIUM POC
5.5 Jun 01

Remote code/prompt injection in NousResearch Hermes Agent through 0.12.0 stems from improper neutralization in the _comp

CVE-2026-10220 MEDIUM POC
5.5 Jun 01

Remote injection in NousResearch Hermes Agent through version 2026.4.30 allows unauthenticated attackers to manipulate t

CVE-2026-9366 MEDIUM POC
5.5 May 24

Code injection in NousResearch hermes-agent 2026.4.23 allows remote unauthenticated attackers to inject and execute arbi

CVE-2026-9354 MEDIUM POC
5.5 May 24

Improper output escaping in NousResearch Hermes Agent versions up to 2026.4.16 allows remote unauthenticated attackers t

CVE-2026-9353 MEDIUM POC
5.5 May 24

Remote injection vulnerability in NousResearch hermes-agent versions up to 2026.4.23 enables unauthenticated attackers t

CVE-2026-10224 MEDIUM POC
5.5 Jun 01

Uncontrolled resource consumption in NousResearch hermes-agent (all versions through 2026.4.30) allows remote unauthenti

This site uses cookies essential for authentication and security. No tracking or analytics cookies are used. Privacy Policy