Back to skill

Security audit

Pipeworx dnd5e

Security checks for vulnerabilities and agentic risk

Overview

This D&D lookup skill is purpose-aligned, but its MCP setup asks users to run an unpinned npm package through npx, which is broader local execution authority than the reference lookup needs.

Review before installing. The D&D query behavior appears straightforward, but use a pinned and reviewed mcp-remote version if possible, avoid unattended first-time npm execution, and run the MCP bridge with minimal access to local files, credentials, and environment variables.

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:46
Finding

Unpinned Third-Party npm Package Is Downloaded and Executed Automatically

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 46-54
Vulnerability Type: Unpinned third-party dependency execution
Risk Level: High

Vulnerable Code

json
{
  "mcpServers": {
    "pipeworx-dnd5e": {
      "command": "npx",
      "args": ["-y", "mcp-remote@latest", "https://gateway.pipeworx.io/dnd5e/mcp"]
    }
  }
}

Technical Analysis

The documented MCP configuration invokes npx with the mutable package specifier mcp-remote@latest. The -y option suppresses interactive confirmation, allowing npm to download and execute whichever release is associated with the latest distribution tag at invocation time.

Because the package version and integrity digest are not pinned, the effective executable payload can change after this skill has been reviewed. The project also provides no lockfile, vendored dependency, provenance verification, or checksum with which to validate the downloaded code.

This is a supply-chain exposure rather than evidence that the current mcp-remote package is malicious. Exploitation would require compromise or malicious modification of the package, its publishing account, the npm distribution path, or a transitive dependency.

Attack Path

  1. A user installs or copies the MCP configuration documented by the skill.
  2. The agent or MCP host launches the configured npx command.
  3. npm resolves the mutable mcp-remote@latest package reference.
  4. The -y option permits installation and execution without user confirmation.
  5. An attacker who controls a future package release, publishing account, or relevant dependency introduces malicious code.
  6. The malicious package code executes locally under the account running the agent or MCP host.

Impact Assessment

Successful exploitation could provide arbitrary code execution with the operating-system privileges of the user running npx. Within those privileges, malicious package code could ac ...[truncated 496 chars]

Remediation
View remediation

Remediation Suggestions

  1. Replace mcp-remote@latest with an exact, reviewed package version rather than a mutable distribution tag.
  2. Use a lockfile and integrity hashes to pin and verify all transitive dependencies.
  3. Review the selected package version and its dependency tree before deployment.
  4. Prefer a locally installed or vendored MCP client whose source and artifacts have been verified.
  5. Verify npm package provenance and signatures where supported.
  6. Avoid unattended first-time installation. Remove -y where practical so unexpected package resolution requires explicit approval.
  7. Run the MCP client in a sandbox or restricted service account with minimal filesystem, credential, environment-variable, and network access.
  8. Establish an explicit dependency-update process in which new versions are reviewed and tested before the pinned version is changed.
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 37)May include surrounding context.

Example: look up Fireball

bash
curl -s -X POST https://gateway.pipeworx.io/dnd5e/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"get_spell","arguments":{"index":"fireball"}}}'