Security audit

marketing-calendar

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent marketing-planning skill that may search the web for missing brand details and create a local HTML report, with no evidence of hidden code, credential use, destructive actions, or persistence beyond the report file.

Install this if you want an automatic Chinese marketing-calendar report generator. Be aware that incomplete brand information may be supplemented with web search, and the generated campaign plan will be saved locally as marketing_calendar_report.html, so avoid entering confidential brand strategy unless that is acceptable.

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

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The instruction to automatically use WebSearch when brand data is incomplete causes external data access without a clear user-facing notice or consent step. This can expose user context or sensitive brand-planning inputs to external services and may also introduce untrusted third-party data into the generated output.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The trigger conditions are broad and include common marketing-related phrases, which can cause the skill to activate in conversations where the user did not explicitly request this specific workflow. Because the skill is also configured to 'not ask, not confirm, directly execute,' unintended activation can lead to unanticipated processing, web searches, and file generation.

Natural-Language Policy Violations

Medium

Confidence: 95% confidence
Finding: The instruction defines the assistant role entirely in Chinese and implicitly requires responses in Chinese, without offering the user a language preference. This can violate language/locale policy when a skill mandates a specific language absent documented justification or opt-in.

Missing User Warnings

Low

Confidence: 89% confidence
Finding: The skill instructs creation of a local HTML report file without informing the user in advance. While lower risk than external network access, unexpected file creation can surprise users, clutter the environment, or write potentially sensitive business content into a persistent artifact they did not knowingly request.

Missing User Warnings

Low

Confidence: 88% confidence
Finding: The instructions direct the skill to generate and write `marketing_calendar_report.html`, which is a file-write operation. In this markdown file, there is no warning that a local file will be created or modified, nor any caution about where it will be saved.

Static analysis

No suspicious patterns detected.