daily-briefing

Pass

Audited by Gen Agent Trust Hub on Jun 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to execute several local commands. It uses date to capture the current timestamp, icalBuddy to retrieve calendar events, and osascript to run AppleScripts located in .claude/skills/ for accessing system emails and reminders.
  • [EXTERNAL_DOWNLOADS]: The skill specifies a requirement for ical-buddy, directing the user to install it via the Homebrew package manager using brew install ical-buddy. This is a well-known utility for accessing macOS calendar data from the command line.
  • [PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection because it processes untrusted data from external sources (emails and calendar events) that are later analyzed by the agent.
  • Ingestion points: Data enters the context from the outputs of scan_inbox.scpt, list_tasks.scpt, and icalBuddy, which can contain arbitrary text from emails or calendar invites.
  • Boundary markers: There are no explicit delimiters or instructions provided to the agent to disregard instructions potentially embedded within the retrieved data.
  • Capability inventory: The skill possesses Bash tool capabilities, which allow it to execute commands and scripts on the local system.
  • Sanitization: The instructions do not define any sanitization or validation logic for the content retrieved from the inbox or calendar before it is summarized by the model.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 30, 2026, 04:01 AM
Security Audit — agent-trust-hub — daily-briefing