ClawHub
Back to skill

Security audit

easypost

Security checks for vulnerabilities and agentic risk

Overview

The skill coherently supports social-media publishing, discloses browser-session reuse, and requires an explicit request before publishing automatically.

Review the EasyPost CLI and extension permissions before connecting logged-in accounts; their implementation is outside this artifact. Prefer the fill-and-review workflow and authorize automatic posting only for intended content and platforms. Use reviewed installation versions where possible. Claims of zero platform-enforcement risk are unverified.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The supplied chunk is an installation command, not an implementation of browser automation or social-platform publishing. It contains no visible content handling, Chrome interaction, or publishing operations. Installation could be a prerequisite, but it is the entire supplied behavior rather than a supporting step within an implemented publishing workflow. The behavior of the downloaded CLI and installed skill cannot be established from this chunk.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
90% confidence
Finding

The installation example invokes an unpinned npm package through npx skills and references a skill repository without an immutable revision. If followed, compromised or unexpectedly changed upstream content could execute installation-time code or introduce malicious skill instructions; however, this is an optional installation example, not an MCP server declaration or evidence of malicious behavior.

Content

No source excerpt is available for this finding.