Back to skill

Security audit

v0-cli

Security checks for vulnerabilities and agentic risk

Overview

This skill is mostly a normal v0.dev automation reference, but it asks users to rely on an unaudited, unpinned third-party CLI and defaults created projects to public visibility.

Install only if you are comfortable auditing or pinning the referenced @hlongvu/v0-cli package yourself. Use --privacy private for sensitive projects, avoid sending secrets or confidential source material to v0.dev, and provide V0_API_KEY only to the command that needs it.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:14
Finding

Unpinned Third-Party CLI Installation Creates a Supply-Chain Risk

Content
View full analysis
[options] ``` Or if installed globally (`npm install -g @hlongvu/v0-cli`): ```bash v0 [options] ``` ### Technical Analysis The documentation directs users to globally install and execute `@hlongvu/v0-cli` without specifying an exact package version or verifying its integrity. An unversioned npm installation resolves to the registry's current package release, meaning the code installed by users can change after this Skill has been reviewed. The audited project contains only `SKILL.md`; it does not include the referenced `src/index.js`, a `package.json`, or a dependency lockfile. Therefore, the implementation, transitive dependencies, installation scripts, and handling of the required `V0_API_KEY` cannot be independently audited from this artifact. This does not establish that the referenced package is currently malicious. It creates a supply-chain exposure under which a compromised publisher account, malicious future release, registry compromise, or package substitution could deliver executable code to users following the documented installation procedure. ### Attack Path 1. An attacker compromises the npm publisher account, publishing workflow, or another component of the package's supply chain. 2. The attacker publishes a malicious release under the existing `@hlongvu/v0-cli` package name. 3. A user or automated agent follows the Skill documentation and runs `npm install -g @hlongvu/v0-cli`. 4. Because no exact version or integrity value is specified, npm retrieves the current compromised release. 5. Malicious lifecycle scripts may execute during installation, or malicious CLI code may execute when the user invokes `v0`. 6. The malicious package runs with the privileges of the installing or invoking acco ...[truncated 712 chars]
Remediation
View remediation
``` 2. Prefer a project-local installation over a global installation and invoke it through a pinned package script or local executable. 3. Include the CLI implementation, `package.json`, and lockfile in the reviewed artifact so its code and complete dependency graph can be audited. 4. Verify package provenance through npm provenance attestations, the expected publisher identity, and a trusted registry configuration. 5. Record and verify package integrity hashes in a lockfile or another authenticated dependency-verification mechanism. 6. Disable npm lifecycle scripts during installation where compatible with the package: ```bash npm install --ignore-scripts ``` 7. Run the CLI under a least-privileged account and provide `V0_API_KEY` only to the specific process that requires it. 8. Document the expected package version, publisher, source repository, and release-verification procedure so users can detect package substitution or unauthorized releases. ]]>
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The create command defaults to public, but the documentation does not prominently warn that generated projects and associated chat content may be publicly visible. This creates a substantial risk that automation or users will unintentionally publish sensitive project details, internal designs, or proprietary prompts simply by using the default behavior.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill instructs users to send natural-language prompts and follow-up chat messages to v0.dev but does not clearly disclose that this content leaves the local environment and is transmitted to a third-party service. In an agent/automation context, users may include proprietary code, internal URLs, credentials, or other sensitive business data in prompts, creating an avoidable data exfiltration risk through inadequate warning and consent.

Content

No source excerpt is available for this finding.