The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the scrapling package from PyPI and uses scrapling install to download necessary browser binaries for its stealth and dynamic fetchers.
[COMMAND_EXECUTION]: The installation process involves executing shell commands such as pip install "scrapling[all]" and scrapling install. Additionally, the skill provides a CLI interface (scrapling extract ...) that involves executing commands with user-provided URLs.
[DATA_EXFILTRATION]: The skill's primary function is to perform network requests to external, non-whitelisted domains to fetch content. While this is the intended purpose, it represents a data transmission vector.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from external websites into the agent's context.
Ingestion points: Content is retrieved from external URLs via scrapling extract (CLI) and Fetcher, DynamicFetcher, or StealthyFetcher (Python).
Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore instructions embedded in the scraped content.
Capability inventory: The skill has network access and the ability to write output to the local file system (e.g., output.md, output.json).
Sanitization: There is no mention of sanitizing or escaping the scraped content before it is presented to the agent.

scrapling