The Agent Skills Directory

[DATA_EXFILTRATION]: The wrapper script scrapling_fetch.py explicitly disables SSL certificate verification by setting verify=False in the fetch_http function call to Fetcher.get. While intended to address certificate compatibility issues on certain systems, this practice significantly increases the risk of man-in-the-middle (MITM) attacks during network operations.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and process arbitrary content from the web, which creates a surface for indirect prompt injection attacks.
Ingestion points: The scrapling_fetch.py script and the Fetcher API classes in SKILL.md retrieve content from external, untrusted URLs provided as input.
Boundary markers: The skill does not implement delimiters or specific instructions to the agent to ignore or isolate potentially malicious instructions embedded in the scraped data.
Capability inventory: The skill possesses network access capabilities and returns external data to the agent context. It does not contain functions for arbitrary file writing or shell command execution based on the scraped content.
Sanitization: There is no evidence of HTML sanitization, script filtering, or instruction escaping performed on the fetched content before it is passed to the agent.
[EXTERNAL_DOWNLOADS]: The script scrapling_install.sh performs an unversioned installation of the scrapling[all] package from the Python Package Index (PyPI). It also executes scrapling install, which triggers the download of Chromium and other browser binaries required for dynamic rendering and stealth fetching.
[COMMAND_EXECUTION]: The skill requires the execution of a setup script scrapling_install.sh that modifies the system environment by installing Python packages and browser dependencies.

scrapling