The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from external sources without sanitization or boundary markers. * Ingestion points: Untrusted data enters the agent context via Gmail message content (use_app_gmail) and web search results (web_search) as described in SKILL.md steps 3 and 4. * Boundary markers: There are no delimiters or instructions to ignore embedded commands when processing this external content. * Capability inventory: The skill has the ability to send emails (send_email_to_user), send SMS messages (send_sms_to_user), and write files to the /home/workspace/Records/Briefings/ directory. * Sanitization: No sanitization, escaping, or validation of external content is performed before interpolation into the final briefing.
[COMMAND_EXECUTION]: The skill executes a local Python script to query task information. Evidence: execution of python3 /home/workspace/Skills/morning-briefing/scripts/tasks_query.py in SKILL.md.

morning-briefing