ClawHub
Back to skill

Security audit

Douban Hot Cn

Security checks for vulnerabilities and agentic risk

Overview

This skill appears non-destructive, but it is materially misleading because it presents itself as a live Douban monitor while only returning static mock movie and book data.

Review before installing. This skill does not appear to steal data or modify your system, but you should treat its results as demo/static sample content unless the publisher updates it to fetch and label real Douban data. It may also trigger on broad movie-related prompts.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
95% confidence
Finding

The skill advertises live Douban hot-topic monitoring, ratings, and review capabilities across movies, books, and music, but the file only documents mock/example outputs and explicitly allows simulated data when APIs are unavailable. This can mislead users or downstream agents into treating fabricated or stale information as real external data, which is a trust and integrity issue even if it is not direct code execution.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger phrases include very broad everyday terms such as '豆瓣', '电影', and 'douban', which may activate the skill during ordinary conversation unrelated to the user’s actual intent. Overbroad activation can cause unintended routing, irrelevant responses, and accidental disclosure or use of this skill when another tool or plain conversation was intended.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill metadata and behavior materially diverge: it claims to provide live Douban trending movies/books/music plus ratings and reviews, but the implementation only returns static mock movie and book data and omits music and reviews entirely. This is dangerous because downstream users or agents may make decisions based on fabricated or stale content while believing it is authoritative current data.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The docstrings state that the functions fetch current Douban hot/in-theater data, but they actually return hardcoded lists. This mismatch can mislead maintainers, reviewers, or calling agents into treating the output as live external data, creating integrity and trust issues even though there is no code execution or direct system compromise.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The user-facing output strings are hard-coded in Chinese for both movie and book modes, and the sample data is also entirely Chinese-language oriented. This creates a language/locale constraint without offering opt-in, fallback, or documentation that the skill is intentionally region-specific.

Content

No source excerpt is available for this finding.