The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses uv run to execute local Python scripts (login.py, batch_publish.py, mcp_server.py) which drive the browser automation logic.
[EXTERNAL_DOWNLOADS]: The skill instructions include playwright install chromium, which downloads browser binaries from Microsoft's official Playwright distribution, a well-known and trusted service.
[CREDENTIALS_UNSAFE]: The skill stores browser session state and authentication cookies in a local state.json file to maintain the user's login session across automation runs, which is a standard implementation for persistent browser automation.
[PROMPT_INJECTION]: The skill processes untrusted text from local chapter files, creating an indirect prompt injection surface.
Ingestion points: Chapter content read from .md and .txt files in local project directories.
Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore embedded commands within chapter text.
Capability inventory: The skill utilizes subprocess execution for scripts, browser automation for publishing, and file system access for archiving successfully uploaded chapters.
Sanitization: No mention of sanitization or filtering for potentially malicious instructions within the processed chapter content.

fanqie-publish