Security audit

Content Writer

Security checks for vulnerabilities and agentic risk

Overview

This is a scoped marketing-writing skill that reads product/brand source material and writes draft files, with no hidden execution, persistence, or publishing behavior found.

Install only if you want a brand-side long-form marketing draft workflow. Be aware that generic writing prompts may invoke it, so provide a clear brief, product name, source location, and desired language; review generated drafts before any publication, especially for customer names, internal facts, and product claims.

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 (8)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The trigger list contains very broad terms such as “draft,” “content brief,” and common Chinese writing requests like “写文章” and “写稿,” which can match many ordinary user intents outside this skill’s narrow purpose. This increases the chance of incorrect auto-invocation or routing, causing the agent to produce brand-side promotional content when the user wanted something else, including neutral analysis or unrelated writing assistance.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The trigger list includes very broad everyday phrases such as '写文章', '写稿', '起草', and 'draft', which can match many ordinary user requests unrelated to this specific marketing-writing skill. This can cause unintended activation, routing user tasks into a brand-marketing workflow when they expected a different tool, increasing the chance of mis-scoped output, policy bypass in downstream orchestration, or accidental processing of inappropriate inputs.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The read_when list contains very broad triggers such as '写文章', '写稿', 'draft', and 'content brief', which can cause this skill to activate for many generic writing requests outside its intended scope. That overmatching can silently route users into a constrained workflow that reads project files and applies brand-specific rules, increasing the chance of unintended file access or inappropriate handling of unrelated tasks.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: This section applies a mandatory separate gate specifically for English drafts, while the skill metadata and body also mix Chinese and English usage, but it does not explicitly offer the user a language-choice flow or explain a policy basis for the language-specific treatment. That creates a natural-language locale policy concern because the skill imposes differentiated handling by language without documented opt-in.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: This reference file presents all guidance exclusively in Chinese and does not offer any language or locale opt-in. Under the policy for natural-language violations, forcing a specific language without user choice can be a locale-policy issue unless the constraint is clearly justified.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: This markdown template is entirely written as a Chinese-language output format and implicitly requires responses in Chinese. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is explicitly justified.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 75% confidence
Finding: 标题明确标注为“中文说明”，且全文均以中文固定描述该技能，没有说明是否支持其他语言或由用户选择语言。按规则，若技能在自然语言层面强制特定语言而无用户选择，可能构成语言/区域策略违例。
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 88% confidence
Finding: This markdown file contains user-facing instructions exclusively in Chinese, and there is no indication that the skill is region-specific or that users may opt into another language. Under the language/locale policy check, forcing a specific language without opt-in is a natural-language policy concern.
Content: No source excerpt is available for this finding.