wechat-markdown-editor

Warn

Audited by Gen Agent Trust Hub on Apr 28, 2026

Risk Level: MEDIUMREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The file scripts/vendor/wechat-md/src/utils/languages.ts dynamically imports JavaScript code from a remote CDN (https://cdn-doocs.oss-cn-shenzhen.aliyuncs.com) at runtime to load syntax highlighting definitions. This allows for the execution of arbitrary remote code if the CDN is compromised.
  • [EXTERNAL_DOWNLOADS]: The downloadFile function in scripts/vendor/wechat-md/src/images.ts fetches content from any URL provided in the markdown source. This creates a Server-Side Request Forgery (SSRF) risk and allows the skill to pull data from untrusted external sources.
  • [COMMAND_EXECUTION]: The skill instructions in SKILL.md require the execution of shell commands using bun or npx to run the conversion logic, which performs file system modifications including reading, writing, and renaming files.
  • [DATA_EXFILTRATION]: The combination of arbitrary file reads (input markdown) and outbound HTTP requests (image downloads) provides a technical pathway for data exfiltration if the processed markdown contains specifically crafted URLs.
  • [PROMPT_INJECTION]: The skill processes untrusted markdown content without explicit boundary markers or instructions for the agent to ignore embedded commands, which could lead to indirect prompt injection if the output is further processed by the agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 28, 2026, 02:02 AM
Security Audit — agent-trust-hub — wechat-markdown-editor