ClawHub
Back to skill

Security audit

rough-cut

Security checks for vulnerabilities and agentic risk

Overview

The skill coherently creates editable videos from your footage, with a routine dependency risk in its command-line fallback.

Use a reviewed, pinned Reelary version for the command-line fallback, index only footage you intend the agent to use, and review the draft before exporting. The companion setup skill is not bundled in this artifact.

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
  • 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 (2)

MCP server referenced without pinned version: 'npx -y reelary'.

Medium
Category
MCP Rug Pull
Confidence
97% confidence
Finding

The fallback invokes an unversioned npm package through npx -y, permitting automatic installation and execution of a changing package release without confirmation. A compromised release could execute with the agent's local permissions and access private footage or other accessible files; the local video-editing context increases the potential exposure, although there is no evidence of malicious intent. This is a CLI supply-chain risk, not specifically an MCP server configuration issue.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx -y reelary'.

Medium
Category
MCP Rug Pull
Confidence
97% confidence
Finding

The second unversioned npx -y reelary invocation exposes the export workflow to the same automatic package execution and supply-chain compromise risk. A malicious package release could access files or execute commands with the agent's privileges rather than merely export video. This is another occurrence of the same underlying weakness, not evidence of a separate attack or malicious author intent.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.