midday-checkin
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script
patterns.pyto perform pattern detection on task and meeting data. It also invokes an external scriptmemory.pyfrom a sibling skill (supermemory) to maintain AI continuity. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the processing of untrusted meeting record data.
- Ingestion points: Meeting records located in the
Records/Meetings/directory are ingested during Phase 2 of the skill execution and by thepatterns.pyscript. - Boundary markers: Absent. The skill does not implement delimiters or instructions to ignore embedded commands within the meeting records.
- Capability inventory: The skill has the capability to send SMS messages to the user, write briefing files to the workspace, and execute local scripts.
- Sanitization: Absent. No validation, filtering, or escaping is performed on the content of meeting records before they are processed by the agent.
Audit Metadata