meetings-prep

Pass

Audited by Gen Agent Trust Hub on Jul 18, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local shell commands to facilitate its functionality. It uses osascript to retrieve events from the Apple Calendar, gog for calendar lookups, and a minutes utility for searching meeting history. These commands are used for their intended administrative purposes.
  • [DATA_EXFILTRATION]: While the skill accesses sensitive user data such as calendar events and meeting transcripts, the analysis confirms that this data remains within the local environment. There are no patterns indicating that this information is transmitted to external or unauthorized domains.
  • [PRIVILEGE_ESCALATION]: The skill includes a chmod 600 command when saving preparation files. This is identified as a security best practice, as it ensures that sensitive meeting briefs are only readable by the owner of the process, preventing unauthorized local access.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from meeting transcripts and calendar descriptions (Phase 2). Although this creates an ingestion surface for potential indirect prompt injections, the skill's logic is focused on summarization and synthesis, minimizing the risk of the agent executing malicious instructions embedded in the transcripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 18, 2026, 11:03 AM
Security Audit — agent-trust-hub — meetings-prep