agent-self-scheduling
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEPERSISTENCEINDIRECT_PROMPT_INJECTION
Full Analysis
- [PERSISTENCE]: The skill provides templates for establishing recurring agent tasks via cron and systemd timers. These mechanisms allow agent processes to run on a recurring schedule and survive system reboots, which is the primary stated purpose of the skill.
- [INDIRECT_PROMPT_INJECTION]: The instructions describe workflows for processing external data (e.g., email summarization) on a schedule, which presents an indirect prompt injection surface.
- Ingestion points: External data enters the agent context via recurring tasks described in SKILL.md, such as the email monitoring example.
- Boundary markers: The provided command examples do not include specific delimiters or 'ignore' instructions to isolate untrusted external content.
- Capability inventory: The skill documentation references the use of pi run, codex exec, and hermes cron to execute agent logic.
- Sanitization: No explicit sanitization or validation of external content is mentioned in the provided scheduling templates.
Audit Metadata