The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions and templates for executing the sau CLI tool for account management and content uploading. The Python template (scripts/examples/kuaishou_cli_template.py) correctly uses subprocess.run with argument lists and shlex.quote for display, mitigating shell injection risks.\n- [EXTERNAL_DOWNLOADS]: The skill references the installation of patchright and Chromium via a well-known mirror (https://npmmirror.com/mirrors/playwright), which is a standard practice for browser automation tools in certain environments.\n- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it processes untrusted user data for titles and descriptions.\n
Ingestion points: User-supplied metadata (--title, --desc, --note) enters the agent context via the CLI contract instructions in references/cli-contract.md.\n
Boundary markers: Absent; the instructions do not specify delimiters for user-provided text.\n
Capability inventory: The skill performs subprocess command execution and file operations (cookie management) as documented in scripts/examples/.\n
Sanitization: Examples use standard shell quoting, but the skill relies on the agent to sanitize input before invoking the CLI.

kuaishou-upload