ClawHub
Back to skill

Security audit

SilicaClaw Network Config

Security checks for vulnerabilities and agentic risk

Overview

This skill is a bounded local-network configuration helper that clearly discloses its ability to inspect and change SilicaClaw runtime visibility settings.

Before installing, understand that this skill can help change whether a local SilicaClaw node is local-only, LAN-visible, or visible on a wider preview network, and whether public discovery is enabled. Use it only when you intend to manage those visibility settings.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Unrestricted Tool Access

Medium
Category
Excessive Agency
Content
It will not:

- execute arbitrary code or untrusted external content
- access unknown remote endpoints outside the documented local workflow
- manage wallets, private keys, or blockchain signing
- silently publish broadcasts as part of network configuration
Confidence
80% confidence
Finding
Skill grants unrestricted tool access without appropriate constraints. An agent with unfettered tool access can perform arbitrary actions including file modification, network requests, and code execution.

Natural-Language Policy Violations

Medium
Confidence
78% confidence
Finding
The manifest includes a natural-language reference named "owner_dialogue_cheatsheet_zh", which indicates Chinese-language guidance is built into the skill package. There is no accompanying indication in this file that language selection is optional or limited to a justified region-specific context, so this may impose a locale-specific experience without user opt-in.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The listed trigger phrases such as “为什么发不出去” are broad and could appear in many unrelated support conversations, yet the file says they are defaultly interpreted as checking network mode and public discovery state. The section does not provide scope constraints or negative examples to distinguish these phrases from general troubleshooting requests.

Static analysis

No suspicious patterns detected.