status-check
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONPERSISTENCE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data sources which could theoretically contain malicious instructions designed to influence the agent's behavior during the status report generation.
- Ingestion points: The skill reads background shell job logs, subagent outputs, monitors, and task lists (SKILL.md).
- Boundary markers: Absent. The instructions do not define specific delimiters or instructions to ignore embedded prompts within the logs or tracker files.
- Capability inventory: The skill uses scheduled-prompt tools (
CronCreate,CronList,CronDelete) but explicitly prohibits merging, pushing, rerunning, or editing (SKILL.md). - Sanitization: Absent. The skill reads raw output lines from jobs and task lists without specified filtering or escaping.
- [PERSISTENCE]: The skill utilizes platform-specific scheduled-prompt tools to create recurring tasks. While this is the intended functionality for periodic reporting, it establishes a mechanism that continues execution after the initial user session concludes. The skill includes built-in limits such as an automatic 7-day expiration and a requirement to stop once no open items remain.
Audit Metadata