Back to skill

Security audit

linkdigest-wechat-article-reader

Security checks for vulnerabilities and agentic risk

Overview

The skill is a disclosed LinkDigest-based WeChat article reader, with remote processing and billing clearly described, though its bundled helper script supports broader LinkDigest inputs than the skill trigger describes.

Install only if you are comfortable sending article links or share text to linkdigest.dev under a paid API key. Do not use it for internal, private, login-only, or sensitive links, because the artifact says results are cached by URL and may be reused for later requests to the same public article. The publisher should ideally add input validation or clearer manifest scope for the broader helper script.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Tp4

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The skill is described as a WeChat-article reader, but the documented behavior indicates broader generic URL/share-text digestion and additional modes/features outside that narrow purpose. This mismatch is dangerous because users and policy systems may authorize the skill under a limited-use assumption while it can actually send arbitrary user-provided content or other links to a third-party service, increasing data-exfiltration and billing risk.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding

The skill explicitly requires environment variables, shell execution, and outbound network access, but it does not declare any tool scope or allowed-tools restrictions. This increases the attack surface because a host agent may grant broader execution than users expect, making it easier for a modified or future version of the skill to misuse shell/network capabilities or access secrets.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 57)May include surrounding context.

没有 Python 时直接调 HTTP:

bash
curl -sS -X POST https://linkdigest.dev/api/v1/digest \
  -H "Authorization: Bearer $LINKDIGEST_API_KEY" -H "Content-Type: application/json" \
  -d '{"url": "https://mp.weixin.qq.com/s/xxxxxxxx", "format": "markdown"}'
# 返回 202 时:curl -sS "https://linkdigest.dev/api/v1/digest/<jobId>?wait=20&format=markdown" -H "Authorization: Bearer $LINKDIGEST_API_KEY"

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest frames this skill specifically as a 微信公众号文章提取 tool for mp.weixin.qq.com articles, including full article Markdown and OCR of embedded images. However, the command-line interface description explicitly states it reads links from 小红书/抖音/TikTok/YouTube/X/公众号文章, indicating the actual code supports a much broader set of sources than the manifest describes.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The manifest describes extracting WeChat article content, OCR text from article images, and related metadata/links. The code adds optional capabilities such as 爆款拆解, translation, and a transcript mode explicitly optimized for long videos, which are materially broader than reading a WeChat article into Markdown.

Content

No source excerpt is available for this finding.