The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the wacli package via the npm package manager (npm install -g wacli), introducing a third-party dependency necessary for its core functionality.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute wacli commands for sending messages, syncing history, and searching conversations, providing the agent with the ability to interact with the local shell.
[INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted data by reading WhatsApp message history and search results, creating a surface for indirect prompt injection. \n
Ingestion points: WhatsApp message content retrieved via wacli history, wacli search, and wacli sync in SKILL.md. \n
Boundary markers: The instructions do not define specific delimiters or guardrails to wrap the external message content before processing. \n
Capability inventory: The agent has access to Bash, Read, and Write tools, enabling command execution and file system modifications. \n
Sanitization: No explicit sanitization or filtering of the message content is mentioned before the agent processes or summarizes it.

whatsapp