Back to skill

Security audit

Agent Ready by WorkPods

Security checks for vulnerabilities and agentic risk

Overview

The scanners report a benign skill with no executable scripts and one disclosed concern: its examples run an unpinned npm package that could change between uses.

The verdict is benign, but benign does not mean risk-free. Before installing or running the examples, consider whether you trust the package to run with your account’s permissions; using a reviewed, pinned version reduces the risk of unexpected changes.

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
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 (2)

MCP server referenced without pinned version: 'npx -y @workpods/agent-ready'.

Medium
Category
MCP Rug Pull
Confidence
98% confidence
Finding

The page-check command uses npx -y with an unversioned npm package, allowing newly resolved releases and their installation scripts to execute with the invoking user's permissions without an installation confirmation. Although this is a CLI example rather than an MCP server invocation, the supply-chain risk is real: a compromised or malicious future release could access local files and credentials, beyond the website-auditing task.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx -y @workpods/agent-ready'.

Medium
Category
MCP Rug Pull
Confidence
98% confidence
Finding

The crawl command also executes an unpinned npm package through npx -y, so its executable code can change without any modification to the skill. The scanner's MCP-specific wording is inaccurate for this CLI example, but a compromised release could still execute arbitrary code with local user permissions; limiting the crawl or respecting robots.txt does not restrict that code's privileges.

Content

No source excerpt is available for this finding.