Back to skill

Security audit

ssai-dailymoments

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed content-and-image generator with local history and MiniMax API use, but I found no hidden, destructive, or purpose-mismatched behavior.

Install only if you want scheduled generation of marketing posts and are comfortable using a MiniMax or compatible image API. Review the cron schedule, API costs, provider data handling, and the permanent local history file; avoid putting sensitive personal, student, or business details into prompts.

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
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill is configured to trigger on very common phrases like “朋友圈” and “发朋友圈”, which can appear in ordinary conversation unrelated to an explicit request to run the automation. This can cause unintended execution, including generation of content, access to local history, and calls to an external image API without a sufficiently deliberate user action.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill states that all generated history is retained permanently, but it does not present this as a clear user-facing warning or obtain consent. Permanent retention increases privacy risk because generated content may reflect user preferences, schedules, or business activity and can accumulate indefinitely in a local file.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill instructs the system to send prompts and related content to a third-party image-generation API, but the description does not clearly warn users that their content leaves the local environment. This creates a transparency and privacy issue, especially if prompts may encode user-specific or business-specific information.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

The documented external endpoint shows that the skill depends on a third-party service for image generation. While not inherently malicious, this is a true data egress path and becomes more dangerous because the skill is designed for automatic and scheduled execution, which can repeatedly send content without per-use user review.

Content

Scanner excerpt · SKILL.md (reported line 89)May include surrounding context.

md
### API配置(用户需自行填写)
- **API Key**:`请填入你的 MiniMax API Key`
- **接口地址**:https://api.minimaxi.com/v1/image_generation
- **模型**:image-01
- **宽高比**:3:4(竖版,适配朋友圈)
- **response_format**:base64(避免URL失效问题)

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

The example code embeds the API key in a variable and demonstrates direct use of a bearer token for outbound requests. Even as sample code, this encourages insecure secret handling practices and reinforces that the skill will transmit data to an external provider.

Content

Scanner excerpt · SKILL.md (reported line 147)May include surrounding context.

md
import base64

api_key = "请填入你的 API Key"
url = "https://api.minimaxi.com/v1/image_generation"
headers = {
    "Content-Type": "application/json",
    "Authorization": f"Bearer {api_key}"

External Transmission

Medium
Category
Data Exfiltration
Confidence
93% confidence
Finding

The Python example performs a live POST request to an external API, confirming that skill data is transmitted off-platform. In this skill context, the transmission is expected for image generation, but it is still security-relevant because it exposes prompts and potentially sensitive operational metadata to a third party.

Content

Scanner excerpt · SKILL.md (reported line 162)May include surrounding context.

md
"seed": 1001
}

resp = requests.post(url, headers=headers, json=data, timeout=30)
result = resp.json()

if result["base_resp"]["status_code"] == 0: