The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references and suggests the installation of the Bun runtime via its official installation script (curl | bash) if the environment check fails. This is a common setup procedure for tools requiring the Bun execution environment.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied markdown and HTML content to format and publish articles. This creates an attack surface where malicious instructions embedded in the source files could potentially influence the agent's behavior during the conversion or publishing process.\n
Ingestion points: Reads and parses user-provided files in scripts/wechat-article.ts, scripts/wechat-browser.ts, and scripts/md-to-wechat.ts.\n
Boundary markers: There are no explicit boundary markers or system instructions used to encapsulate the user content to prevent the model from obeying embedded instructions.\n
Capability inventory: The skill has capabilities to perform network requests (to WeChat APIs), write files, and execute shell commands for browser automation.\n
Sanitization: The skill does not perform sanitization of the input text specifically to strip out potential prompt injection attempts before processing.\n- [COMMAND_EXECUTION]: To support browser-based publishing, the skill executes system-level commands (such as osascript on macOS, xdotool on Linux, and powershell on Windows) to simulate keystrokes and manage the system clipboard. These are necessary to bypass web-based editor restrictions when inserting rich content like HTML and images.\n- [DYNAMIC_EXECUTION]: The skill uses the Chrome DevTools Protocol (CDP) to interact with the WeChat Official Account web interface. This involves using Runtime.evaluate to execute custom JavaScript within the browser context to automate tasks like filling forms, uploading images, and clicking menu items.\n- [DATA_EXFILTRATION]: The skill retrieves WeChat API credentials (WECHAT_APP_ID and WECHAT_APP_SECRET) from local .env files. These credentials are transmitted to the official WeChat API endpoint (api.weixin.qq.com) solely for the purpose of authenticating article publishing requests.

aiset-post-to-wechat