ClawHub
Back to skill

Security audit

Raspberry Pi First Setup / 树莓派首次配置

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Raspberry Pi setup guide that only installs its helper tool after explicit approval and keeps device validation read-only.

Before installing, confirm you are comfortable with an npm registry check and a global jishubuddy install on your operator computer. Review the displayed version and command, disable JishuBuddy telemetry before launch if desired, and only approve device configuration or storage writes when you have verified the target device and data impact.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (5)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding
The skill promises Raspberry Pi setup and validation but mainly provides guidance plus a permissioned path to query npm and install JishuBuddy on the operator machine. That mismatch can mislead users about what actions will actually occur and introduces undeclared external network access to the npm registry, which is a security-relevant side effect in a skill that otherwise appears advisory.

Ae1

High
Category
analysis-evasion
Content
`<skill-directory>` to the absolute directory containing this `SKILL.md`,
Confidence
100% confidence
Finding
Referenced artifact was not completely inspected

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
| Use when | A Raspberry Pi needs imaging preparation, first-boot setup, or initial connectivity checks. |
| Delivers | A setup checklist and evidence-backed first-boot status, including checks still awaiting confirmation. |
| Requires | Exact board/setup details; a supported operator computer running JishuBuddy for live SSH or serial checks. |
| Does not | Erase storage without approval or treat application installation as validated device setup. |

## Use this skill when
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.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
| Use when | A Raspberry Pi needs imaging preparation, first-boot setup, or initial connectivity checks. |
| Delivers | A setup checklist and evidence-backed first-boot status, including checks still awaiting confirmation. |
| Requires | Exact board/setup details; a supported operator computer running JishuBuddy for live SSH or serial checks. |
| Does not | Erase storage without approval or treat application installation as validated device setup. |

## Use this skill when
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.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
devices.
3. Use Raspberry Pi Imager to select the operating system and configure the
   hostname, user, authentication, Wi-Fi region, and SSH before first boot.
4. Prefer SSH keys over password authentication. Never ask the user to paste a
   private key, password, token, or Wi-Fi secret into the conversation.
5. Use a reliable SD card for initial setup. Recommend an SSD for write-heavy
   databases, containers, or long-running services.
Confidence
80% 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.

Static analysis

No suspicious patterns detected.