Back to skill

Security audit

social-media-search

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a disclosed social-media search helper, but it uses unsafe user-input command templates that could become command injection if run through a shell.

Install only if you are comfortable with a skill that automatically opens Chinese social-media sites and sends your search terms there. Avoid using untrusted or specially crafted search strings until the publisher changes the instructions to require explicit slash-prefixed commands and safe argument or URL encoding.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Error
Location
SKILL.md:55
Finding

Shell Command Injection Through Unsanitized Search Terms

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The Douyin trigger list includes broad natural-language phrases like '搜索抖音', '抖音搜索', and '在抖音搜索' without requiring a strict command prefix or other activation boundary. That makes accidental invocation more likely during ordinary conversation, which can cause unintended browser navigation to external sites and execution of automated actions based on user text.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The Xiaohongshu triggers are similarly broad and can match ordinary user phrasing rather than deliberate tool use. Because the skill then opens a browser and types into a known page element automatically, ambiguous activation can lead to unintended external requests and automation on a third-party site.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The Bilibili trigger set includes generic phrases such as 'B站搜索' and '哔哩哔哩搜索' without activation boundaries, creating a risk of collisions with normal discussion. In a tool-enabled agent, this can unexpectedly launch browser navigation with user-provided query content, expanding the attack surface for prompt-triggered or accidental actions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

Natural-language policy review applies to all file types. The skill description is specified only in Chinese and the trigger/output conventions throughout the file are Chinese-specific, with no opt-in, language selection, or documented region-specific justification.

Content

No source excerpt is available for this finding.