知乎发帖-谷歌浏览器

• Outcome: Review
• Latest audit: May 29, 2026 at 9:27 AM UTC
• Version:
  0.0.3

Overview

The skill is open about publishing Zhihu articles, but it can automatically post under a logged-in account without a fresh confirmation step.

Install only if you want an agent to publish real Zhihu column articles from your logged-in Chrome profile. Prefer using explicit preview or draft wording unless you are ready to post, and consider editing the skill to require confirmation before --submit.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: The trigger list includes broad natural-language phrases such as 帮我发帖 and 自动发知乎 that can match ordinary conversation and unintentionally activate a skill that performs real external actions. Because this skill is configured to publish content, accidental invocation can lead directly to unauthorized posting from an already logged-in account.

Missing User Warnings

High

• Confidence: 98% confidence
• Finding: The workflow explicitly says the default for normal requests is to skip checks and go straight to --submit, meaning the agent will perform a real publication without a dedicated confirmation barrier. In a session-attached browser context, this can cause irreversible posting under the user's account due to misunderstanding, prompt injection, or accidental trigger matches.

VirusTotal

65/65 vendors flagged this skill as clean.