web-scraper-data-miner

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions suggest using the bash tool to run scraping operations with libraries such as Playwright and Cheerio.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it is designed to process untrusted data from the open web.
  • Ingestion points: Untrusted web content retrieved via crawl_tool or bash scripts in SKILL.md.
  • Boundary markers: Absent; there are no instructions to use delimiters or specifically warn the agent against following instructions found within the scraped data.
  • Capability inventory: The skill utilizes bash for command execution, which could be leveraged if an injection occurs.
  • Sanitization: While the instructions mention cleaning and validating extracted data, they do not include measures to sanitize the content for potential malicious instructions before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 05:33 PM
Security Audit — agent-trust-hub — web-scraper-data-miner