ClawHub
Back to skill

Security audit

SwarmClaw

Security checks for vulnerabilities and agentic risk

Overview

The skill is a markdown-only multi-agent orchestration package, but it describes broad agent execution, file operations, deployments, shared memory, and webhooks without clear user approval or containment guidance.

Review this skill carefully before installing. It should only be used in tightly scoped workspaces and should require explicit user confirmation before file writes, deletes, deployments, webhook use, or production-impacting API calls. The package is not malicious based on the inspected artifacts, but its orchestration claims are broader than its safety instructions.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill advertises file operations and deployment automation capabilities but provides no warning, approval requirements, or guardrails for potentially destructive actions. In an agent-orchestration context, these capabilities can lead to unintended file modification, deletion, or deployment changes if users assume the agent is safe for autonomous execution.

Content

No source excerpt is available for this finding.