ClawHub
Back to skill

Security audit

Social media specs

Security checks for vulnerabilities and agentic risk

Overview

The skill provides coherent social-media formatting guidance, with an ordinary software-trust consideration around its suggested validation commands.

You can use the supplied reference tables without running the CLI. If you use it, review and pin the package version and run it with limited access to sensitive files. The external package's executable code was not verified in this review.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

The recommended npx commands can download and execute third-party code without a user warning.

Medium
Category
Output Handling
Confidence
91% confidence
Finding

The skill recommends third-party code execution as a routine validation step without explaining that npx may download and run a package with local permissions. Some npx configurations prompt before installation, so execution without any warning is not guaranteed, but the skill itself supplies no trust or approval guidance; this overlaps with the three unpinned-command findings rather than representing a separate exploit.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx social-media-specs'.

Medium
Category
MCP Rug Pull
Confidence
95% confidence
Finding

This is a CLI invocation, not an MCP server, but the underlying unpinned-execution concern is valid: npx can download and execute a mutable release of social-media-specs with the agent's permissions. A compromised package release could access local files or credentials; using executable third-party code merely to count caption length creates an avoidable trust boundary.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx social-media-specs'.

Medium
Category
MCP Rug Pull
Confidence
95% confidence
Finding

Although the detector incorrectly calls this an MCP server, this command executes an unpinned npm CLI to validate media dimensions. A malicious or compromised release could run arbitrary code with the agent's permissions; the harmless example arguments do not restrict what the package can do.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx social-media-specs'.

Medium
Category
MCP Rug Pull
Confidence
95% confidence
Finding

The unversioned npx command can execute a changing third-party package, even though it is a CLI rather than an MCP server. If that package is compromised, it could access the supplied ad copy and other resources available to the execution environment; nothing in the provided skill demonstrates malicious package behavior.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.