data-scraper-agent
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill implements automated task execution via GitHub Actions workflows to run Python-based scraping and AI enrichment logic on a cron schedule.- [EXTERNAL_DOWNLOADS]: The scraper is designed to download content from external web pages and REST APIs as part of its data collection functionality, utilizing common libraries like
requestsandplaywright.- [PROMPT_INJECTION]: The skill architecture is susceptible to indirect prompt injection (Category 8) because it ingests data from external websites and incorporates it into LLM prompts without robust sanitization or delimiters. \n - Ingestion points:
scraper/sources/my_source.py(fetching external content). \n - Boundary markers: None identified in the
_build_promptfunction to isolate external data from instructions. \n - Capability inventory: Scraped content influences AI-generated scores and summaries, which are stored in the user's database (Notion/Sheets). \n
- Sanitization: No sanitization or escaping of scraped text is performed prior to inclusion in prompts.
Audit Metadata