The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the md2wechat command-line interface to perform all its primary functions, including article conversion and metadata inspection.
[CREDENTIALS_UNSAFE]: The skill interacts with sensitive environment variables (WECHAT_APPID, WECHAT_SECRET) and configuration files. The instruction to use md2wechat config show could lead to the unintentional display of these secrets in the agent's output window.
[EXTERNAL_DOWNLOADS]: The skill includes capabilities to download remote images from arbitrary URLs via the download_and_upload command, which is an external data fetch operation.
[DATA_EXFILTRATION]: The skill is designed to upload local Markdown content, HTML, and images to external WeChat servers and potentially AI service providers, representing a controlled data transfer to third-party platforms.
[PROMPT_INJECTION]: The skill handles untrusted external data which creates a surface for indirect prompt injection:
Ingestion points: Reads local Markdown files and fetches remote images.
Boundary markers: There are no instructions for the agent to use delimiters or ignore instructions embedded within the Markdown files.
Capability inventory: The agent has permissions to execute shell commands, access the network, and read local files.
Sanitization: The workflow does not specify any validation or sanitization of the input Markdown before it is processed by AI-related commands like humanize or write.

md2wechat