Back to skill

Security audit

Clawhub Package Mcp

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed local-memory MCP integration with consent and deletion guidance, though users should verify the external npm package and non-ASCII documentation link before installing.

Before installing, verify the npm package name, pinned version, registry integrity, and the GitHub documentation account, then configure the MCP server with a dedicated storage location and only store memories after explicit consent. Treat stored facts and thread memory as persistent local data and delete the memory files or use wipe/forget commands when no longer needed.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Ae4

Medium
Category
analysis-evasion
Confidence
91% confidence
Finding

The document contains mixed-script content, most notably a Cyrillic-looking repository owner name in the support/docs URL. Mixed-script Unicode can enable homograph confusion, making users believe they are visiting a trusted project or domain when the identifier is visually deceptive. In installation documentation, this is more dangerous because users are being directed to fetch software and support resources.

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 100)May include surrounding context.

text

New releases are reviewed and tested before they are promoted — do not
switch to a newer version without checking its changelog. Run the MCP
server with least-privileged local access (its own storage path only).

## Tools

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The support section lists only a Telegram bot handle with a Russian-language name (@ViBomemorybot), and the docs link also points to a repository under a Cyrillic-named account. This can imply a locale-specific support/documentation expectation without explicitly offering users a language choice or stating that support is region-specific.

Content

No source excerpt is available for this finding.