skill-event-calendar
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Python scripts to provide core functionality. It runs
scripts/lunar.pyfor specialized calendar math andskills/shared/scripts/calendar_ops.pyfor event synchronization. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from a reference file, creating a surface for potential instruction injection. \n
- Ingestion points: The skill reads event and festival data from
references/events-china.md. \n - Boundary markers: The instructions lack explicit delimiters or 'ignore embedded instructions' warnings to isolate the ingested data from the agent's command context. \n
- Capability inventory: The skill can execute shell commands (
python3) and read sensitive user profile files (identity.md,preferences.md). \n - Sanitization: There is no documented step for validating or sanitizing the markdown content before it is parsed into the agent's reasoning loop.
Audit Metadata