Back to skill

Security audit

Pipeworx microlink

Security checks for vulnerabilities and agentic risk

Overview

This skill appears purpose-aligned, but it asks users to run a mutable npm package automatically and sends URL-based work to a third-party remote service with limited disclosure.

Review before installing. Avoid using this with confidential, internal, or token-bearing URLs, and prefer a pinned mcp-remote version or a sandboxed low-privilege MCP environment.

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

Error
Location
SKILL.md:22
Finding

Unpinned npm Dependency Is Automatically Downloaded and Executed

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 22–24
Vulnerability Type: Unpinned executable third-party dependency
Risk Level: High

Vulnerable Code

json
"command": "npx",
"args": ["-y", "mcp-remote@latest", "https://gateway.pipeworx.io/microlink/mcp"]

Technical Analysis

The documented configuration invokes npx with the -y option to download and execute mcp-remote@latest automatically. The mutable latest tag does not identify a fixed, audited package version and provides no integrity constraint. Consequently, the code executed by users can change after this Skill has been reviewed without any corresponding modification to the repository.

The -y option suppresses the normal installation confirmation, increasing the likelihood that a changed or compromised release will execute without user review. This creates a supply-chain trust boundary involving the npm registry, the package publisher, and the package's transitive dependencies.

The configuration also passes an external MCP endpoint to the downloaded package. The remote service can change independently, but the repository provides no evidence that the endpoint itself is malicious; the confirmed vulnerability is the automatic execution of the unpinned npm dependency.

Attack Path

  1. An attacker compromises the npm publisher account, package release process, registry distribution path, or a relevant dependency.
  2. The attacker publishes a malicious version under the latest tag of mcp-remote.
  3. A user applies the documented MCP configuration.
  4. npx -y resolves the mutable tag, downloads the malicious release, and executes it without an interactive installation confirmation.
  5. The malicious package runs with the permissions and environment available to the process that launched the Agent or MCP client.
  6. It may access files, environment variables, credentials, or network resources available to that account, subject to operating-system and sandbox ...[truncated 712 chars]
Remediation
View remediation

Remediation Suggestions

  1. Replace mcp-remote@latest with an explicitly pinned and reviewed version, such as mcp-remote@x.y.z.
  2. Verify the selected release and its transitive dependencies before deployment.
  3. Use lockfiles and npm integrity hashes where the surrounding installation mechanism supports them.
  4. Prefer a controlled installation step over automatic execution through npx -y; removing -y can at least preserve an interactive confirmation boundary.
  5. Consider mirroring the approved package artifact in a trusted internal registry and restricting dependency retrieval to that source.
  6. Run the MCP process in a sandbox or dedicated low-privilege account with minimal filesystem, credential, and network access.
  7. Establish a deliberate update and re-audit process rather than following a mutable distribution tag.
  8. Document what information is transmitted to https://gateway.pipeworx.io/microlink/mcp and restrict sensitive URLs or content from being submitted unless necessary.
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 (1)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill routes requests to a remote MCP endpoint and exposes tools that operate on user-supplied URLs, but the description does not warn users that submitted URLs and potentially fetched page content will be sent to an external Microlink service. This creates a real transparency and data-handling risk because users may unknowingly transmit sensitive internal URLs, tokens in query strings, or private page content to a third party.

Content

No source excerpt is available for this finding.