agent-self-scheduling

Pass

Audited by Gen Agent Trust Hub on Jul 26, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a tutorial for setting up agent persistence and scheduling. It demonstrates the use of standard system utilities like cron and systemd, as well as platform-specific scheduling features for Hermes, Claude Code, and Codex. No malicious intent or suspicious commands were identified in the provided documentation.
  • [PROMPT_INJECTION]: The skill describes patterns where an agent processes data from external sources (such as summarizing emails via the 'himalaya' CLI), which creates an inherent surface for indirect prompt injection.
  • Ingestion points: External data enters the agent context from sources like email clients or external status checks.
  • Boundary markers: The illustrative examples do not demonstrate the use of delimiters or 'ignore' instructions to isolate untrusted content.
  • Capability inventory: The documentation references agent environments that may have access to bash execution, reading, and editing tools.
  • Sanitization: Guidance for sanitizing or validating external content before it is processed by the LLM is not explicitly addressed in the provided snippets.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 26, 2026, 03:28 PM
Security Audit — agent-trust-hub — agent-self-scheduling