The Agent Skills Directory

[DATA_EXFILTRATION]: The initialization playbook (scripts/initialize.md) contains instructions for an automated background agent to perform reconnaissance on a user-specified URL, capturing screenshots and extracting visible API endpoints and integration details.- [COMMAND_EXECUTION]: The tunneling guide (scripts/tunnel.md) provides instructions to execute shell commands that expose local network ports to the internet using third-party services like Cloudflare or ngrok, which can lead to unauthorized external access if not carefully managed.- [EXTERNAL_DOWNLOADS]: The skill fetches software and configurations from well-known technology services and official package registries, including the vendor's own SDK and official tools from Cloudflare.- [DATA_EXFILTRATION]: The knowledge base module (references/knowledge-base.md) is designed to ingest and transmit local file content to a remote API for search indexing.- [PROMPT_INJECTION]: The skill manages conversational flows and chat widgets that process untrusted user input, creating a surface for indirect prompt injection, though it utilizes schema-based validation for state management.

waniwani-sdk