Back to skill

Security audit

Code Fix

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent debugging helper, but it includes unguarded commands that can delete dependencies and execute unpinned packages from external registries.

Install only if you are comfortable with a debugging skill that may suggest shell commands and project edits. Before following its package-manager cleanup or npx examples, confirm the project path, prefer local lockfile-pinned binaries, avoid automatic execution in sensitive repositories, and approve destructive cleanup or dependency installation explicitly.

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:195
Finding

Unpinned Third-Party Package Retrieval and Execution

Content
View full analysis
# Installed version tree npm outdated # Packages that can be updated npx synp --source-file # yarn.lock ↔ package-lock.json conversion check # Clear cache rm -rf node_modules && npm install ``` ### Technical Analysis The documented debugging workflow instructs the agent to execute `npx tsc` and `npx synp` without requiring an audited, lockfile-pinned local package. If the requested executable is unavailable locally, `npx` may retrieve a package from the configured package registry and execute its code. The effective code executed can therefore depend on mutable external registry content rather than code reviewed as part of this Skill. The `npx tsc` invocation also identifies only a binary name rather than explicitly ensuring that the binary comes from the intended, project-declared `typescript` dependency. The `npx synp` command similarly permits retrieval and execution of an unpinned utility. In addition, `rm -rf node_modules && npm install` deletes the current dependency tree and performs a new installation. Dependency installation can execute package lifecycle scripts such as `preinstall`, `install`, and `postinstall`. If the manifest, lockfile, registry configuration, or a resolved package is malicious or compromised, this operation can cause arbitrary local code execution. ### Attack Path 1. A JavaScript or TypeScript debugging request activates the Skill. 2. The agent follows the technical-stack reference and invokes an `npx` command when the corresponding trusted local executable is absent. 3. `npx` resolves and downloads an unpinned package from the configured pac ...[truncated 1136 chars]
Remediation
View remediation
tsc --version ``` Direct local, lockfile-backed execution remains preferable. 4. Replace general restoration through `npm install` with deterministic lockfile installation where appropriate: ```bash npm ci --ignore-scripts ``` 5. Enable lifecycle scripts only when they are necessary and after reviewing the relevant dependency scripts. Do not automatically run installation commands against untrusted repositories. 6. Require explicit user authorization before deleting `node_modules`, downloading packages, or executing dependency lifecycle scripts. 7. Validate registry configuration and lockfile integrity before installation. Use an approved registry, retain integrity hashes, and incorporate dependency scanning into the workflow. ]]>
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Rp1

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The skill recommends running npx tsc --version without pinning a package/version. npx may fetch and execute code from the registry when the binary is not already present locally, which creates a supply-chain execution risk in the agent context. Because this is a debugging skill that normalizes shell execution, the unsafe example is more likely to be followed automatically or semi-automatically.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
97% confidence
Finding

The skill suggests npx synp --source-file, which can cause npx to download and execute the latest package from the registry if synp is not installed locally. That exposes users or agents to arbitrary third-party code execution via dependency confusion, compromise of the upstream package, or unexpected version changes. In a skill intended to troubleshoot projects, this is especially risky because such commands may be run in sensitive repositories or developer environments.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill instructs destructive cleanup commands such as removing node_modules and cleaning npm cache without requiring confirmation, impact assessment, or backup guidance. While common in debugging, these commands can delete large dependency trees, disrupt offline or reproducible environments, and trigger costly reinstalls or state loss when executed automatically by an agent. The debugging context makes this more dangerous because the workflow encourages direct shell remediation steps during incident handling.

Content

No source excerpt is available for this finding.