Security audit

Bilibili

Security checks for vulnerabilities and agentic risk

Overview

This is a benign Markdown-only Bilibili content-strategy skill with no executable code, permissions, persistence, or hidden data handling.

Install if you want Bilibili-specific video-script and danmu resonance advice. Be aware that it uses Chinese platform terms by design and may be best suited to Bilibili or Chinese-language creator workflows.

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
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 (5)

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The trigger list contains broad, common creator-discussion phrases such as Bilibili strategy, script, resonance points, and optimization terms that could cause the skill to activate outside a clearly intended invocation context. In an agent ecosystem, overly broad triggers can lead to unintended routing, prompt collisions, or opportunistic interception of user requests that were not meant for this skill.

Natural-Language Policy Violations

Low

Confidence: 77% confidence
Finding: The README uses multiple Chinese terms and platform-specific phrasing such as '弹幕触发点', '槽点', '梗点', '共鸣点', and 'B站-native' without stating that the skill is intended for Chinese-speaking users or offering an opt-in language choice. This can create a language/locale policy issue if users are implicitly forced into a specific language context.

Natural-Language Policy Violations

Low

Confidence: 84% confidence
Finding: This markdown file includes example prompts and outputs that assume Chinese-language usage and Bilibili-specific Chinese terms such as 发弹幕, 吐槽, 槽点, 梗点, and 名场面. Because the examples do not indicate that Chinese output is optional or user-selected, they may imply a fixed language/locale behavior without opt-in.

Natural-Language Policy Violations

Low

Confidence: 76% confidence
Finding: The skill exposes keywords and triggers in both English and Chinese, but the manifest does not clarify whether the skill adapts to the user's preferred language or locale. Because policy violations include forcing a specific language without opt-in, the lack of a documented language-choice mechanism creates a mild natural-language policy concern.

Natural-Language Policy Violations

Low

Confidence: 82% confidence
Finding: The standard output format hard-codes Chinese reaction categories such as “吐槽 / 梗 / 共情 / 站队 / 高能 / 名场面” as the output schema. While the skill is Bilibili-specific, the file does not explicitly offer users an option to receive output in another language or localized format, which can be a language/locale policy concern.

Static analysis

No suspicious patterns detected.