Back to skill

Security audit

Pipeworx artic

Security checks for vulnerabilities and agentic risk

Overview

The skill’s museum-data purpose is coherent, but its MCP setup runs an unpinned npm package automatically, which creates a supply-chain review risk.

Before installing, consider pinning mcp-remote to a reviewed exact version, avoiding automatic install confirmation where practical, and running the MCP process with limited filesystem and environment access. The API behavior appears focused on public museum data, but the setup depends on trust in the Pipeworx gateway and the npm package resolution path.

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:48
Finding

Unpinned npm Package Is Automatically Downloaded and Executed

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 48-56
Vulnerability Type: Unpinned third-party dependency execution
Risk Level: Medium

Vulnerable Code

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

Technical Analysis

The MCP configuration invokes npx with the automatic-confirmation option (-y) and the mutable dependency specifier mcp-remote@latest. When the MCP server starts, npm may download and execute whichever package release is associated with the latest distribution tag at that time.

Because the effective package code can change after this Skill has been reviewed, the local configuration does not provide a reproducible or auditable dependency boundary. Compromise of the npm package, its publisher account, or its release process could cause subsequently launched installations to execute attacker-controlled package code.

No evidence establishes that the current package is malicious. The vulnerability is the unsafe use of a mutable, automatically executed dependency rather than a confirmed malicious payload.

Attack Path

  1. An attacker compromises the npm package, its publisher credentials, or the package release process.
  2. The attacker publishes a malicious package version and assigns it to the latest distribution tag.
  3. A user enables or starts the documented MCP configuration.
  4. npx -y downloads the version currently referenced by mcp-remote@latest without requesting interactive confirmation.
  5. npm executes the downloaded package under the account that launched the MCP client.
  6. The malicious package can access files, environment variables, credentials, and network resources available to that account.

Impact Assessment

Successful exploitation could provide arbitrary code execution with the privileges of the user ...[truncated 494 chars]

Remediation
View remediation

Remediation Suggestions

  1. Replace mcp-remote@latest with an exact, reviewed package version, such as mcp-remote@x.y.z.
  2. Manage the dependency through a committed lockfile and require integrity verification using the package manager's recorded cryptographic hash.
  3. Preinstall the reviewed dependency during a controlled build or deployment phase rather than downloading it automatically when the MCP server starts.
  4. Remove -y where practical so unexpected installation or package-resolution changes require explicit approval.
  5. Use a trusted internal npm registry or allowlist and monitor the package for ownership changes, unexpected releases, and security advisories.
  6. Run the MCP process in a restricted environment with minimal filesystem access, sanitized environment variables, limited network access, and no administrative privileges.
  7. Re-audit and deliberately update the pinned version instead of inheriting changes through a mutable distribution tag.
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)

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 39)May include surrounding context.

Look up Grant Wood's American Gothic (artwork ID 6565):

bash
curl -s -X POST https://gateway.pipeworx.io/artic/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"get_artwork","arguments":{"id":6565}}}'