Executive Summary
In November 2026, the Git project disclosed and patched CVE-2026-111760, a critical integer-truncation vulnerability allowing malicious repositories to execute arbitrary code on developer machines during recursive clones. An attacker crafts an unusually long path that bypasses Git's safety check; Git then truncates it to a name resolvable as a .git file within a submodule, enabling replacement with a symbolic link pointing to attacker-controlled hooks or configuration. Code execution occurs automatically when Git processes those hooks. Patches were released across Git versions 2.50.2 through 2.56.1, with GitHub, GitLab, and GitHub Enterprise applying fixes immediately alongside five additional CVEs covering path traversal, stack corruption, heap overflow, and arbitrary file read/write.
This incident underscores escalating risk in software supply-chain attack surfaces, where developer tooling itself becomes the vector. As CI/CD pipelines and AI-assisted coding workflows increase automated recursive clone operations, a single unpatched Git binary across a development fleet or cloud build environment can silently compromise entire codebases, secrets stores, and downstream artifacts before any runtime security control is engaged.
Why This Matters Now
Recursive Git clones are a default, automated behavior in modern CI/CD pipelines, GitHub Actions, and AI-assisted development tools like GitHub Copilot. With six CVEs patched simultaneously and Git for Windows lagging behind at disclosure, millions of developer endpoints and cloud build agents remain exposed during the update window. Supply-chain attackers exploiting this flaw can inject malicious hooks silently, compromising build artifacts and secrets before any runtime or network control can intervene.
Attack Path Analysis
An attacker crafts a malicious Git repository containing an extremely long path that exploits CVE-2026-111760, an integer truncation flaw, causing Git to write a symbolic link named '.git' that redirects submodule processing to an attacker-controlled directory containing malicious hooks; when a victim developer or CI/CD pipeline recursively clones the repository, the hooks execute arbitrary attacker code on the cloning host. From that initial foothold, the attacker escalates privileges using the execution context of the pipeline or developer workstation, moves laterally across connected cloud environments and repositories, establishes command-and-control via outbound callbacks, and may exfiltrate secrets, source code, or credentials — ultimately disrupting software supply chains and downstream deployments.
Kill Chain Progression
This analysis maps confirmed threat intelligence to the full cloud kill chain to show where defensive gaps would emerge as an attack progresses.
Initial Compromise
Description
The attacker publishes or controls a malicious Git repository containing an artificially long path that exploits CVE-2026-111760 (integer truncation in Git's index handling). When a victim performs a recursive clone including submodules, Git writes a symbolic link named '.git' pointing to an attacker-controlled directory holding malicious hook scripts. Git executes these hooks automatically during clone operations, achieving arbitrary code execution on the cloning host — whether a developer workstation or a CI/CD runner. The attack requires no credentials and no interaction beyond the clone command.
Related CVEs
CVE-2026-111760
CVSS 9An integer truncation vulnerability in Git's index handling allows an attacker to craft a repository with an extremely long path that bypasses safety checks and is written to disk under a shorter attacker-controlled name such as .git, enabling replacement of a submodule's .git file with a symbolic link pointing to a directory containing malicious hooks or configuration, resulting in arbitrary code execution during recursive clones.
Affected Products:
Git Project Git – < 2.50.2, < 2.51.3, < 2.52.1, < 2.53.1, < 2.54.1, < 2.55.1, < 2.56.1
GitHub GitHub Enterprise Server – < 3.19.15, < 3.20.11, < 3.21.9, < 3.22.4
Exploit Status:
no public exploitCVE-2026-113408
CVSS 7.5When cloning an empty repository over Git protocol v2, a malicious server can inject a crafted branch name containing path traversal sequences, which with the files reference backend can allow writes outside Git's reference folder during a subsequent branch update.
Affected Products:
Git Project Git – < 2.50.2, < 2.51.3, < 2.52.1, < 2.53.1, < 2.54.1, < 2.55.1, < 2.56.1
Exploit Status:
no public exploitCVE-2026-114467
CVSS 6.5Merging a tag with more than two GPG signatures causes stack memory corruption in Git, which can result in a denial-of-service condition and may affect hosting services that perform automated merges on behalf of users.
Affected Products:
Git Project Git – < 2.50.2, < 2.51.3, < 2.52.1, < 2.53.1, < 2.54.1, < 2.55.1, < 2.56.1
Exploit Status:
no public exploitCVE-2026-114469
CVSS 5.5A binary patch that expands to a very large size can bypass Git's approximately 1 GiB patch size limit, triggering an out-of-bounds read and crashing Git.
Affected Products:
Git Project Git – < 2.50.2, < 2.51.3, < 2.52.1, < 2.53.1, < 2.54.1, < 2.55.1, < 2.56.1
Exploit Status:
no public exploitCVE-2026-105437
CVSS 7.5An extremely large merge operation involving many conflicts combined with a large conflict-marker-size attribute causes Git to write past the end of an undersized heap buffer, resulting in heap memory corruption and potential code execution on code-hosting servers running git merge-tree.
Affected Products:
Git Project Git – < 2.50.2, < 2.51.3, < 2.52.1, < 2.53.1, < 2.54.1, < 2.55.1, < 2.56.1
GitLab GitLab CE/EE – affected versions running vulnerable Git builds
Exploit Status:
no public exploitCVE-2026-114466
CVSS 7.1Three git fast-import features (rewrite-submodules-from, rewrite-submodules-to, and export-pack-edges) can be abused via a crafted input stream to read arbitrary files or create files and append arbitrary data, unless --allow-unsafe-features is explicitly set.
Affected Products:
Git Project Git – < 2.50.2, < 2.51.3, < 2.52.1, < 2.53.1, < 2.54.1, < 2.55.1, < 2.56.1
Exploit Status:
no public exploit
MITRE ATT&CK® Techniques
Compromise Software Dependencies and Development Tools
Exploitation for Client Execution
Command and Scripting Interpreter
Hijack Execution Flow: Executable Installer File Permissions Weakness
Masquerading: Match Legitimate Name or Location
File and Directory Discovery
Data Manipulation: Stored Data Manipulation
Escape to Host
Potential Compliance Exposure
Mapping incident impact across multiple compliance frameworks.
NIST SP 800-53 Rev 5 – Flaw Remediation
Control ID: SI-2
CISA Zero Trust Maturity Model 2.0 – Software Supply Chain Integrity
Control ID: Applications Pillar – Application Security
NIS2 Directive – Policies on Risk Analysis and Supply-Chain Security
Control ID: Article 21(2)(a) and 21(2)(e)
PCI DSS 4.0 – All System Components Are Protected from Known Vulnerabilities
Control ID: Requirement 6.3.3
DORA (EU Regulation 2022/2554) – ICT Asset Vulnerability and Patch Management
Control ID: Article 8 – ICT Risk Management
ISO/IEC 27001:2022 – Management of Technical Vulnerabilities
Control ID: Annex A 8.8
Sector Implications
Industry-specific impact of the vulnerabilities, including operational, regulatory, and cloud security risks.
Computer Software/Engineering
Git's CVE-2026-111760 supply-chain flaw enables malicious code execution during recursive clones, directly threatening software build pipelines, CI/CD workflows, and developer environments reliant on submodule integrity.
Information Technology/IT
IT organizations using GitHub Actions, GitHub Enterprise Server, and Codespaces face supply-chain compromise risk; unpatched Git versions expose lateral movement and exfiltration vectors across hybrid cloud environments.
Computer/Network Security
Security tooling built on Git-based repositories risks submodule hook hijacking; zero trust segmentation and egress policy enforcement are critical mitigations against post-clone attacker-controlled code execution.
Financial Services
Financial institutions using Git-managed infrastructure-as-code face supply-chain attacks enabling privilege escalation and data exfiltration, triggering PCI 4.0, NIST 800-53, and ZTMM compliance exposure from CVE-2026-111760.
Sources
- Git Patches Flaw That Lets Malicious Repositories Execute Code During Recursive Cloneshttps://thehackernews.com/2026/11/git-patches-flaw-that-lets-malicious.htmlVerified
- Git Security Vulnerabilities Announced - GitHub Bloghttps://github.blog/open-source/git/git-security-vulnerabilities-announced-7/Verified
- Git 2.56.1 Release Announcement - lore.kernel.orghttps://lore.kernel.org/git/xmqqh5hosgw5.fsf@gitster.g/Verified
- GitLab Security Releaseshttps://about.gitlab.com/releases/categories/releases/Verified
- CISA Known Exploited Vulnerabilities Cataloghttps://www.cisa.gov/known-exploited-vulnerabilities-catalogVerified
- NVD - National Vulnerability Databasehttps://nvd.nist.gov/Verified
Frequently Asked Questions
Cloud Native Security Fabric Mitigations and ControlsCNSF
Based on the attack progression modeled above, these are the defensive controls that would constrain each stage.
Aviatrix Zero Trust CNSF is relevant here because the attacker's post-compromise progression depends entirely on unrestricted network reachability between CI/CD environments, cloud workloads, and external infrastructure. By enforcing segmentation, identity-scoped east-west policies, and controlled egress, the fabric would likely reduce the attacker's ability to pivot from an exploited build runner into broader cloud environments and establish persistent outbound callbacks.
Control: Cloud Native Security Fabric (CNSF)
Mitigation: The CNSF fabric would likely constrain the network reachability of the compromised cloning host, reducing its ability to initiate unauthorized connections to attacker infrastructure immediately following hook execution, even though the initial code execution on the host itself may still occur.
Control: Zero Trust Segmentation
Mitigation: Zero Trust segmentation would likely reduce the scope of cloud resources reachable using harvested credentials by constraining which workloads and services the CI/CD runner's identity is permitted to communicate with, making broad credential abuse harder to operationalize across the environment.
Control: East-West Traffic Security
Mitigation: East-west traffic enforcement would likely constrain the attacker's ability to traverse from the build environment into connected production or staging workloads, reducing the blast radius of the initial compromise by limiting which internal cloud destinations the attacker could reach using stolen credentials.
Control: Multicloud Visibility & Control
Mitigation: Multicloud visibility and control would likely expose anomalous outbound connection patterns from the CI/CD workload to unrecognized external destinations, reducing the attacker's ability to maintain an undetected and persistent command-and-control channel across cloud environments.
Control: Egress Security & Policy Enforcement
Mitigation: Egress policy enforcement would likely constrain the attacker's ability to transfer sensitive data to unauthorized external destinations by reducing the set of permitted outbound paths available from cloud workloads hosting secrets, source code, or credential material.
If segmentation, east-west controls, and egress enforcement meaningfully constrained earlier stages, the attacker's reach into artifact repositories and downstream deployment pipelines would likely be reduced, narrowing the scope of supply chain poisoning to a smaller subset of workloads rather than propagating broadly across connected environments.
Impact at a Glance
Affected Business Functions
- Software Development Pipelines and CI/CD Workflows
- Source Code Repository Integrity and Supply Chain Security
- Developer Workstation and Build Environment Security
- Code Hosting and Collaboration Platform Operations
Estimated downtime: 3 days
Estimated loss: N/A
No confirmed data exfiltration reported at time of disclosure. However, the primary risk is arbitrary code execution on developer workstations and CI/CD build servers during recursive git clone operations, which could lead to: compromise of source code repositories and intellectual property, injection of malicious code into build artifacts and software supply chains, theft of developer credentials and access tokens stored in local environments, and lateral movement into enterprise development infrastructure. Organizations relying on automated pipelines that clone untrusted or third-party repositories face the highest exposure risk. The supply-chain blast radius could extend to downstream software consumers if malicious code is injected into distributed build artifacts.
Recommended Actions
Key Takeaways & Next Steps
- • Apply CNSF Egress Security & Policy Enforcement (FQDN filtering, outbound policy) on all CI/CD runner and developer build environments to block unauthorized outbound callbacks initiated by malicious Git hooks before C2 channels can be established.
- • Deploy Cloud Firewall (ACF) with URL and FQDN-based egress controls around build pipelines and cloud-connected developer workstations to detect and deny connections to attacker-controlled infrastructure triggered by hook execution.
- • Enforce Zero Trust Segmentation using identity-based microsegmentation policies to isolate CI/CD runners from production cloud workloads, limiting lateral movement from a compromised build host to connected cloud environments and artifact stores.
- • Enable East-West Traffic Security controls between build environments, container registries, and cloud workload tiers so that anomalous workload-to-workload traffic generated post-compromise is detected and blocked before attackers can pivot into production.
- • Activate Multicloud Visibility & Control to establish traffic baselines for CI/CD pipelines and trigger anomaly alerts on unexpected outbound connections, unusual cloud API calls, or atypical data volumes that may indicate hook-initiated code execution or credential harvesting.



