The Agent Skills Directory

[REMOTE_CODE_EXECUTION] (CRITICAL): The file SKILL.md contains instructions to pipe remote scripts to bash: curl -sSL https://canifi.com/skills/motion/install.sh | bash. This is a confirmed remote code execution pattern from an untrusted source that bypasses security reviews.
[EXTERNAL_DOWNLOADS] (HIGH): The skill references https://canifi.com/install.sh and other scripts from canifi.com, which is not a recognized trusted organization or repository.
[COMMAND_EXECUTION] (MEDIUM): The skill utilizes a custom command-line utility canifi-env for local configuration, which involves executing shell commands to manage environment variables and sensitive credentials.
[PROMPT_INJECTION] (LOW): The skill is vulnerable to indirect prompt injection through task and project data ingestion.
Ingestion points: User-provided task descriptions and project names in the task management examples.
Boundary markers: Absent; user input is interpolated directly without delimiters or 'ignore' instructions.
Capability inventory: The skill has the capability to schedule meetings, allocate work time, and reorganize focus time.
Sanitization: Absent; no evidence of input validation or safety filtering for external content.

motion