The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill directs users to external domains including facebook18.com and multiple Google Sites (sites.google.com/view/...) for system access, scripts, and documentation. These are not verified or well-known services.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external social media platforms which is then used in automated messaging functions.
Ingestion points: Data is extracted via FollowerExtractor.get_followers, FollowerExtractor.search_by_hashtag, and TikTokExtractor.extract_engaged_users (SKILL.md).
Boundary markers: No delimiters or safety instructions are present to prevent the agent from being influenced by malicious content embedded in the scraped usernames, comments, or bios.
Capability inventory: The skill possesses significant output capabilities, including sending bulk WhatsApp messages (MassSender.send_bulk_message), automated Instagram DMs, and comments (EngagementBot.run_campaign).
Sanitization: There is no evidence of sanitization or filtering of extracted external content before it is interpolated into message templates like dm_template or comment_templates.
[COMMAND_EXECUTION]: The skill provides numerous Python snippets that implement high-frequency automated actions (follows, likes, DMs, mass messaging) across WhatsApp, Instagram, and TikTok, which may lead to account restrictions or safety concerns when executed.

whatsapp-mass-sender-group-marketing