The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script located at {user.workspace}/scripts/meeting_prep_assertions.py to verify the integrity of briefings and cron job creation.\n- [COMMAND_EXECUTION]: Schedules automated reminders and post-meeting tasks using the openclaw cron utility, which involves persistent command scheduling on the host system.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted data from email bodies and calendar event descriptions without sanitization.\n
Ingestion points: Calendar event details and Gmail message history (90-day and historical lookbacks).\n
Boundary markers: Absent; the instructions do not specify delimiters or warnings to ignore instructions embedded within the ingested communications.\n
Capability inventory: The skill can execute shell commands (mcporter, openclaw, python3) and deliver data to external messaging platforms.\n
Sanitization: Absent; external content is summarized and reformatted directly into the final briefing output.\n- [DATA_EXFILTRATION]: Transmits sensitive user data, including summarized email threads and meeting notes, to external platforms (WhatsApp and Slack). This is the intended primary purpose of the skill and targets the user's own accounts.\n- [DATA_EXFILTRATION]: Accesses and utilizes sensitive configuration details from ~/executive-assistant-skills/config/user.json, including contact information and workspace paths.

meeting-prep