Description

A security flaw has been discovered in adolfosalasgomez3011 slidev-builder-mcp 2.1.0. This affects the function generateChart of the file src/tools/generateAssets.ts of the component generateAssets Tool. Performing a manipulation of the argument outputDir results in command injection. The attack is only possible with local access. The project was informed of the problem early through an issue report but has not responded yet.

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "5.3",
  "pubDate": "2026-08-08T12:16:46.057Z",
  "pubdate": "2026-08-08T12:16:46.057Z",
  "executiveSummary": "A security flaw has been identified in adolfosalasgomez3011 slidev-builder-mcp version 2.1.0, specifically within the generateChart function of the src/tools/generateAssets.ts file residing in the generateAssets Tool component.\nThis vulnerability manifests as a command injection flaw occurring when untrusted input is passed to the outputDir argument without adequate sanitization or validation.\nSuccessful exploitation of this security issue allows a threat actor to execute arbitrary system commands on the underlying operating system.\nThe risk implication is severe, potentially leading to complete system compromise, unauthorized data access, and lateral movement within the affected environment.\nAnalysis of the attack prerequisites indicates that the exploitation vector requires local access to the system to manipulate the vulnerable argument.\nDespite early notification via an issue report, the project maintainers have not yet responded or provided an official patch to remediate the vulnerability.",
  "technicalDetails": "The vulnerability resides in the slidev-builder-mcp package, specifically affecting version 2.1.0 of the software.\nThe vulnerable component is the generateAssets Tool, with the exact point of failure located in the generateChart function within the src/tools/generateAssets.ts source file.\nThe root cause of the vulnerability is the unsafe handling and concatenation or execution of the outputDir argument, which passes raw, unsanitized input directly into an underlying operating system shell command interpreter.\nBecause input validation and proper parameterization are absent, an attacker can inject malicious shell metacharacters and command sequences into the outputDir parameter.\nThe attack flow proceeds as follows: First, the adversary identifies the local execution context where the generateAssets Tool is invoked. Second, the attacker crafts a malicious payload designed to break out of the intended directory path context and execute arbitrary system commands. Third, the attacker supplies this crafted payload via the outputDir argument to the generateChart function. Fourth, the application processes the input by passing it directly to the system shell. Finally, the injected commands execute with the privileges of the running Node.js process, leading to unauthorized command execution.\nRegarding exploitation requirements, the attack is only possible with local access to the target system.\nAuthentication requirements and specific network exposure vectors are not explicitly defined beyond the requirement for local interaction with the vulnerable module or CLI interface.\nPrivilege requirements depend on the security context of the user executing the slidev-builder-mcp process, but successful command execution inherits those local privileges.\nThe payload behavior involves direct execution of operating system commands, which can facilitate data exfiltration, system manipulation, or further post-exploitation activities."
}