The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs its primary functions by executing the gws binary with various subcommands. This is the intended design for managing Google Meet resources such as conference records, participants, recordings, and transcripts.
[SAFE]: No malicious patterns, such as hardcoded credentials, obfuscation, or unauthorized network activity, were detected. The skill properly delegates authentication and global security configurations to a shared external skill file (../gws-shared/SKILL.md).
[SAFE]: The skill possesses an ingestion surface for external data by retrieving meeting transcripts and notes, which is a necessary feature for its stated purpose.
Ingestion points: External data enters the agent context via CLI outputs from gws conferenceRecords.transcripts and gws conferenceRecords.smartNotes.
Boundary markers: None explicitly defined in the skill instructions.
Capability inventory: Execution of system commands through the gws binary.
Sanitization: Not specified; the skill relies on the underlying tool and agent-level safety filters.

gws-meet