scrapling

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill has a high surface area for indirect prompt injection because it is designed to fetch and process arbitrary web content for an AI agent.\n
  • Ingestion points: Web content is retrieved through multiple tools like get, fetch, and stealthy_fetch as documented in references/mcp.md and scripts/scrape_list.py.\n
  • Boundary markers: The skill does not implement explicit delimiters or system-level instructions to ignore potential commands within the scraped text.\n
  • Capability inventory: The agent has access to networking (via fetchers), file system operations (via script output), and environment commands.\n
  • Sanitization: While the library provides text cleaning and structural conversion (Markdown/HTML), it does not sanitize the content against prompt-based attacks.\n- [EXTERNAL_DOWNLOADS]: Setup involves the scrapling install command which downloads browser binaries (Chromium and Camoufox) required for operation.\n- [COMMAND_EXECUTION]: The skill uses terminal commands for installation, dependency management, and executing scraping scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 07:17 PM