Publish To Mdpage

• Outcome: Pass
• Latest audit: May 28, 2026 · 8:39 AM UTC
• Version: 1.2.0

Overview

This skill does what it says: uploads chosen markdown to md.page to create a shareable link, with no hidden install or persistence behavior found.

Install only if you want an agent to upload markdown to md.page and create shareable links. Review any file, note, report, personal data, confidential material, or credential-like text before publishing; treat the resulting URL as shareable with anyone who has the link.

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Vague Triggers

High

• Confidence: 97% confidence
• Finding: The trigger description is overly broad and can activate on common phrases like 'share this' or 'create a page' without clearly signaling that the content will be sent to an external service. In an agent setting, this increases the chance of accidental exfiltration of sensitive markdown, notes, reports, or file contents to a third party when the user may not understand the privacy implications.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill promotes publishing markdown but does not prominently warn, in the user-facing description, that the content is transmitted to md.page, a third-party external service. This omission undermines informed consent and makes accidental disclosure of sensitive content more likely, especially because the skill is designed for sharing arbitrary markdown and existing files.

External Transmission

Medium

• Category: Data Exfiltration

Content

**POST** `https://md.page/api/publish`

```bash
curl -X POST https://md.page/api/publish \
  -H "Content-Type: application/json" \
  -d '{"markdown": "# Hello\n\nYour content here"}'
```

• Confidence: 93% confidence
• Finding: This instruction explicitly sends user-provided markdown to an external endpoint, creating a real data exfiltration path. Even though the skill includes some advice about scanning for secrets, it still enables transfer of arbitrary content to a third party and is dangerous if used on sensitive reports, notes, or files without robust consent and content classification controls.

External Transmission

Medium

• Category: Data Exfiltration

Content

3. To safely handle special characters (quotes, apostrophes, newlines), use Python to JSON-encode the payload, then pipe it to curl:

```bash
curl -s -X POST https://md.page/api/publish \
  -H "Content-Type: application/json" \
  -d "$(python3 -c "
import json

• Confidence: 92% confidence
• Finding: This second example provides an operational recipe for packaging arbitrary markdown and posting it to md.page, making external transmission easy and repeatable. In context, the ease of use combined with broad activation phrases makes inadvertent publication of confidential content more likely, despite the stated intent being functionality rather than abuse.

Static analysis

No suspicious patterns detected.