ClawHub
Back to skill

Security audit

Huggingface

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward Hugging Face CLI reference skill, but users should be careful with tokens, uploads, and delete commands.

Install only if you want an agent to help operate the Hugging Face CLI. Use a least-privilege Hugging Face token, avoid exposing tokens in commands or logs, review paths before upload, prefer explicit file paths over broad `.` uploads, verify repo visibility, and require confirmation before delete or move operations.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill requires an `HF_TOKEN` and mentions token-based authentication but does not warn users to protect credentials from shell history, logs, screenshots, or accidental inclusion in commands. In agent-assisted workflows, token mishandling can result in account compromise, unauthorized repository changes, or access to private models and datasets.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill documents a destructive `hf repos delete` command without any caution, confirmation guidance, or scoping advice. In an agent setting, this increases the risk of accidental deletion of valuable remote repositories or data, especially if a user asks for broad repo management and the agent treats deletion as a normal operation.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill encourages uploading local files and directories to Hugging Face, including `.` and whole folders, without warning about sensitive data exposure. In an agent context, this can lead to unintentional exfiltration of secrets, proprietary models, credentials, `.git` history, or personal data if the local path is broader than intended or the target repo visibility is misunderstood.

Static analysis

No suspicious patterns detected.