The Agent Skills Directory

[COMMAND_EXECUTION]: The skill documents and facilitates the use of the backend.auth.cmd configuration parameter in ~/.config/himalaya/config.toml, which allows the execution of arbitrary shell commands (e.g., pass show email/imap) to retrieve authentication credentials.
[DATA_EXFILTRATION]: The skill requires access to ~/.config/himalaya/config.toml, a sensitive file path that stores email account configurations and potentially credentials, as detailed in SKILL.md and references/configuration.md.
[DATA_EXFILTRATION]: The skill provides capabilities to send data over the network to external SMTP servers via the himalaya template send and himalaya message write commands, which could be abused to exfiltrate information.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted data from external sources (emails).
Ingestion points: The himalaya message read and himalaya envelope list commands in SKILL.md bring external email bodies and subjects directly into the agent's context.
Boundary markers: No boundary markers or instructions to ignore embedded commands are present in the provided documentation or command examples.
Capability inventory: The skill possesses significant capabilities, including sending emails (himalaya template send), deleting messages (himalaya message delete), and modifying folder structures (himalaya message move).
Sanitization: There is no evidence of sanitization, filtering, or validation of email content before it is processed by the agent.

himalaya