The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from arbitrary websites to extract lead data, which can lead to indirect prompt injection attacks.
Ingestion points: External website content is fetched in scripts/extract_website_contacts.py via fetch_page and converted to markdown.
Boundary markers: The model prompt in scripts/extract_website_contacts.py (lines 211-255) uses a simple WEBSITE CONTENT: header without robust delimiters or instructions to ignore embedded commands within the ingested text.
Capability inventory: The skill possesses the capability to write data to Google Sheets via gspread and perform additional network requests relative to business enrichment.
Sanitization: The skill uses html2text to strip HTML tags, which provides basic cleanup but does not sanitize potential malicious natural language instructions that might be present in the visible text of a webpage.

gmaps-leads