scrapling-mcp

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or security vulnerabilities were detected in the skill files. Analysis confirmed that all external references point to official documentation or well-known repositories.
  • [INDIRECT_PROMPT_INJECTION]: The skill inherently processes untrusted data from the web through its scraping tools, creating a surface for indirect prompt injection. This is an expected risk for a scraping utility.
  • Ingestion points: Web content fetched via MCP tools (fetch_page, fetch_dynamic, fetch_stealthy) and helper scripts (scripts/scrapling_scrape.py, scripts/scrapling_smoke_test.py).
  • Boundary markers: No explicit markers are defined in the scripts to delimit untrusted data from instructions.
  • Capability inventory: The skill facilitates network operations via fetchers and file writing via spider results (to_jsonl).
  • Sanitization: The provided scripts perform extraction but rely on the agent or user for final data validation.
  • [EXTERNAL_DOWNLOADS]: The documentation provides standard instructions for installing the scrapling package and its dependencies from official registries like PyPI.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 12:20 AM