The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructs the agent to automatically and silently install the @marswave/listenhub-cli package using npm install -g. Installing software from unverified third-party sources can lead to the execution of untrusted code.
[COMMAND_EXECUTION]: The skill triggers shell commands (listenhub auth status, listenhub auth login) to manage account status and authentication, which involves interacting with the local system environment and user session data.
[PRIVILEGE_ESCALATION]: The skill requests a global package installation (npm install -g), which typically requires elevated system privileges. It instructs the agent to perform this action 'silently' and to 'never ask the user', effectively bypassing user consent and oversight for privileged operations.
[INDIRECT_PROMPT_INJECTION]: The skill defines a route for content extraction from external URLs (/content-parser), creating a vulnerability surface where remote content could influence agent behavior.
Ingestion points: Remote content extracted from user-provided URLs as defined in the routing table of SKILL.md.
Boundary markers: Absent. There are no instructions to isolate the parsed content from the agent's control logic or to treat it strictly as data.
Capability inventory: The skill possesses the ability to execute shell commands and install software globally.
Sanitization: Absent. There is no mention of sanitizing, filtering, or validating the content retrieved from external URLs before processing.

listenhub-cli