skills/openclaw/skills/scrapling/Gen Agent Trust Hub

scrapling

Pass

Audited by Gen Agent Trust Hub on Mar 19, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill represents a significant surface for indirect prompt injection because its core function is to fetch and process data from untrusted external websites.
  • Ingestion points: The skill uses Scrapling's fetchers (Fetcher, DynamicFetcher, StealthyFetcher) to retrieve content from any URL provided to the agent.
  • Boundary markers: There are no programmed boundary markers or explicit instruction-filtering mechanisms in the provided Python scripts (scrapling_scrape.py and scrapling_smoke_test.py) to prevent the agent from interpreting instructions embedded in scraped HTML or JSON data.
  • Capability inventory: The skill possesses network access and file-system write capabilities (demonstrated in spider-recipes.md), which could be targeted by instructions found in scraped content.
  • Sanitization: The scripts extract content via selectors but do not perform sanitization or escaping of the raw data before passing it back to the agent context.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to install external dependencies required for its operation.
  • Evidence: Documentation in SKILL.md and mcp-setup.md directs the user to install the scrapling and playwright packages from standard registries (PyPI).
  • Context: These downloads target well-known open-source libraries and are consistent with the skill's stated purpose of providing web scraping capabilities.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 19, 2026, 06:54 AM