The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: Installation instructions in SKILL.md recommend a high-risk execution pattern: curl -sSL https://raw.githubusercontent.com/pimalaya/himalaya/master/install.sh | PREFIX=~/.local sh. This executes unverified remote code from a repository that is not part of the trusted vendor list.
[EXTERNAL_DOWNLOADS]: The skill initiates downloads of an installation script from GitHub and retrieves email content and attachments from remote IMAP servers as specified in user configuration.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes untrusted data from external sources.
Ingestion points: Commands such as himalaya message read and himalaya message export in SKILL.md bring external email content into the agent's context.
Boundary markers: The instructions lack markers or warnings to the agent to isolate the email content from its own instructional logic.
Capability inventory: The skill provides the agent with capabilities to send messages (himalaya template send), delete emails, and manage account configurations.
Sanitization: No logic is described for sanitizing or validating the incoming email text before it is processed by the agent.

himalaya