What is the CVSS score of CVE-2026-106764?

CVE-2026-106764 has a CVSS 3.1 base score of 7.5 (High). CVSS vector: CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H. EPSS exploitation probability: 0.7%.

How to fix or mitigate CVE-2026-106764?

Within 24 hours, identify any use of @rog0x/mcp-git-tools in your environment and restrict its exposure to untrusted networks. Within 7 days, review vendor guidance and monitor for a patch; if a patch is not available, implement strict input validation and limit user interaction with untrusted repositories. Within 30 days, apply the vendor-released patch once available and verify that all instances are updated.

@rog0x/mcp-git-tools CVE-2026-106764

EUVD-2026-109095 HIGH

Code Injection (CWE-94)

2026-10-23 mitre

GHSA-jwc4-922r-f2rw

RCE Node.js Code Injection

7.5

CVSS 3.1 · Vendor: mitre

View

Attack ChainAIDerived

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

Access

Lure agent/dev to attacker-controlled repo
Exploit

Supply malicious base branch input
Execution

Trigger flaw in resolveBaseBranch
Impact

Execute arbitrary code on host

Vulnerability Assessment

Weakness	CWE CWE-94 OWASP Top 10:2025 A05:2025 - Injection ASVS 5.0 V1 - Encoding and Sanitization Code injection prevention, sandbox exec MITRE ATT&CK 2 techniques across 2 tactics: Execution, Initial Access The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up. Sign in to read
Exploitation Conditions	Exploitation targets the resolveBaseBranch function of @rog0x/mcp-git-tools 1.0.1 and requires that function to process attacker-controlled input (e.g., a crafted base-branch name or repository the victim's MCP server/agent operates on); per the CVSS vector, user/agent interaction is required (UI:R) and attack complexity is high (AC:H), so the victim must actively invoke the tooling and specific conditions must align. No authentication to the package is required (PR:N). The exact trigger within resolveBaseBranch is not described in the available sources. Limiting factors: the required interaction, the high-complexity conditions, and the package's tiny installed base sharply narrow realistic exploitation.
Impact	The signals are mixed and should be read against a very small deployment footprint. The MITRE-provided CVSS 3.1 vector (AV:N/AC:H/PR:N/UI:R, C:H/I:H/A:H, base 7.5 HIGH) describes a high-impact but non-trivial attack: no authentication is needed (PR:N), yet high complexity (AC:H) and required user/agent interaction (UI:R) mean exploitation is conditional rather than spray-and-pray. No EPSS probability is supplied, there is no CISA KEV listing, and no confirmed public exploit is indicated - so there is no evidence of widespread or automated exploitation. The vuln.today decision label is 'this_week', indicating elevated but not emergency urgency. The dominant risk-reducing factor is prevalence: @rog0x/mcp-git-tools is an obscure individual npm package (an MCP 'git tools' server), so even a true unauthenticated RCE affects a tiny population. Net: treat as a real but narrow issue - relevant mainly to developers/AI-agent operators who actually installed this package, not a broad fleet-wide emergency. Caveats to verify: CWE is N/A and the trigger mechanism of resolveBaseBranch is not described in the sources; the CPE is a junk 'n/a' placeholder; and one reference is a third-party advisory (stillosdigitalholdings.com) that is not independently verified.
Exploit ScenarioDerived	A developer or AI agent runs the @rog0x/mcp-git-tools MCP server and points it at an attacker-influenced repository or supplies attacker-controlled branch input; when resolveBaseBranch processes that input, it leads to arbitrary code execution on the host. The CVSS vector's UI:R and AC:H indicate the victim must invoke the affected tooling and specific conditions must be met, so this is not drive-by. No confirmed working exploit code is indicated in the input, though a GitHub issue reference exists that may contain technical detail.

This Week Schedule remediation this week. Why: HIGH severity (CVSS 7.0-8.9).

No patch listed yet 0.7% exploitation probability (EPSS)

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

Alert me

AnalysisAIv1 – Oct 23, 2026

Arbitrary code execution in @rog0x/mcp-git-tools 1.0.1 is reachable through the package's resolveBaseBranch function when it processes attacker-controlled input, such as a crafted base-branch name or repository handled by a victim's MCP server or agent. Per the assessed CVSS vector the flaw is network-reachable and requires no authentication (PR:N), but attack complexity is high (AC:H) and the user or agent must actively interact with the tooling (UI:R) before the condition aligns; the exact trigger inside resolveBaseBranch is not described in the available sources. No CISA KEV entry or EPSS signal was supplied, and no verified working exploit code is documented - the references include a public GitHub issue thread for the project that should be treated as a public report reference rather than confirmed exploit code. Practical risk is further narrowed by the package's very small deployment footprint, and the overall decision label is this_week rather than immediate emergency action.

RemediationAI

No vendor-released patch was identified at the time of analysis: the available references - the npm package page, GitHub issue #1, and the advisory JSON - do not list a fixed version, so there is no confirmed upgrade target to cite. The most direct action is to remove @rog0x/mcp-git-tools 1.0.1 from any MCP server or agent configuration until the maintainer publishes a fix or closes the issue with a remediation commit, and to treat the npm package page and the GitHub issue thread as the tracking source for that update. Where the package must remain in use, constrain the input path that feeds resolveBaseBranch: do not let an agent or remote caller supply arbitrary base-branch names or repository references (use an explicit allowlist of branch names and repositories), which reduces exposure but does not remove the flaw and will break workflows that depend on dynamic branch resolution. Additionally, run the MCP server under a dedicated low-privilege account with no write access to sensitive repositories or credentials on the host, so a successful trigger cannot escalate to broader code execution; this is a containment measure with operational overhead and does not substitute for a patched release.

Affected VersionsENISA

n/a n/a

Affects your stack

Find out whether the stacks and servers you run are on an affected version.

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

Technical ContextAI · show details

@rog0x/mcp-git-tools is an npm package that exposes git operations as tools to a Model Context Protocol (MCP) server or agent, and the affected function, resolveBaseBranch, is the component that determines/validates a base branch for repository operations. The reported impact is execution of arbitrary code by a remote attacker who can influence the input that resolveBaseBranch resolves (for example a crafted base-branch name or the repository the victim's MCP server operates on); because a model-driven agent may forward externally derived strings into the tool, the trust boundary between the agent's input and the git invocation is the relevant surface. Neither the CWE nor the vulnerability class is supplied in the source data, so the precise root cause (for example command/argument injection versus unsafe path handling) is not confirmed; the function name and the RCE tag suggest an injection-style flaw but this should be treated as inference, not evidence. The assessed vector (AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H) confirms network accessibility and the absence of an authentication barrier, but explicitly requires user/agent interaction and high-complexity preconditions, which materially limits automated mass exploitation. The CPE field is a placeholder (n/a), so no structured version mapping is available beyond the stated 1.0.1 version in the description and the linked npm package page.

Know when Node.js 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

Continue with Google Continue with GitHub

No credit card · 7-day full trial

Severity by source

Vendor (mitre) PRIMARY 7.5 HIGH

AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H

vuln.today AI 7.5 HIGH

Unauthenticated (PR:N) RCE, but an agent/user must invoke the tool on crafted input (UI:R) and the undescribed trigger implies specific conditions, so AC:H; full C/I/A impact from code execution.

3.1 AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H

4.0 AV:N/AC:H/AT:N/PR:N/UI:A/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N

Primary rating from Vendor (mitre).

CVSS VectorVendor: mitre

Attack Vector

Network

Attack Complexity

High

Privileges Required

None

User Interaction

Required

Scope

Unchanged

Confidentiality

High

Integrity

High

Availability

High

DescriptionCVE.org

An issue in @rog0x/mcp-git-tools 1.0.1 allows a remote attacker to execute arbitrary code via the resolveBaseBranch function.

Executive ViewAI

Why It Matters

A high-severity vulnerability in the @rog0x/mcp-git-tools package could allow an attacker to execute arbitrary code when a user or agent interacts with crafted input. This affects only organizations using this specialized tool, and while no active exploitation has been observed, the potential for remote code execution warrants prompt remediation.

Business Impact: Medium

Compensating Controls

No vendor-released patch identified at time of analysis. Interim mitigations include restricting network access to MCP servers, validating repository and branch names, and limiting user interaction with untrusted inputs.

Threat Intelligence Overview

EPSS Score

0.7%

Top 47% exploitable

Prevalence

Rare

obscure individual npm MCP package, negligible production deployment footprint

Data Quality

5 sources

Confidence: Medium

EUVD

NVD

CVE.org

EPSS

GitHub

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 HIGH

CISA KEV No

Public POC No

EPSS 0.7%

Patch available No

Active threat No

Deciding rule

HIGH severity (CVSS 7.0-8.9)

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· derived

Scheduled

Exploitation· derived

None

System Exposure· derived

Open

Automatable· derived

Human Impact

Medium

Priority Score

Low Medium High Critical

KEV: 0

EPSS: +1

CVSS: +38

PoC: 0

Weaponized: 0

Threat intel: 0

Gov advisory: 0

Attack Technique

Code Injection CRITICAL

Code injection occurs when an application accepts user input and passes it directly into a language interpreter or evaluator without proper sanitization.

Remote Code Execution CRITICAL

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