ClawHub
Back to skill

Security audit

asset-controlled-export-pipeline

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed local asset-export and metadata-injection workflow, with no evidence of hidden exfiltration, credential access, persistence, or destructive behavior.

Install only if you intend to manage a local SynomosAI/Ardot-style controlled asset library. Before running the scripts, verify the _asset_index.json paths and URL map so files are written only where expected, and be aware that the optional model mode sends asset metadata prompts to a local service on port 8080.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill advertises a controlled asset export pipeline but declares no permissions while static analysis detected file read, file write, and network-capable behavior. In a workflow handling brand/design assets, those capabilities can enable silent local copying and external exfiltration of sensitive files without an explicit permission boundary, which is especially risky given the governance-heavy language but absence of concrete technical restrictions.

Vague Triggers

Medium
Confidence
77% confidence
Finding
The trigger model is broad enough that ordinary multilingual queries about assets or exports may activate the skill unintentionally. Because this skill is positioned to process and export controlled design assets, accidental invocation could lead to inappropriate access, file generation, or distribution actions in contexts where the user did not intend to run an export workflow.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The manifest description includes many broad trigger phrases such as asset export, controlled assets, VI distribution, copyright metadata, and design asset management without clear scope limits, exclusions, or user-intent boundaries. This can cause the skill to activate in loosely related contexts and handle sensitive brand assets or export workflows when not explicitly intended, increasing the chance of overbroad data handling or misuse.

Static analysis

No suspicious patterns detected.