long-running-jobs
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a robust and secure pattern for managing tasks that exceed the duration of an agent turn. It emphasizes using scheduler-level dependencies and durable filesystem records rather than expensive or fragile in-context polling.
- [COMMAND_EXECUTION]: The provided scripts (scripts/run-and-mark.sh) execute user-specified commands in the background to monitor their completion. This is the intended and documented functionality of the skill, used to capture exit codes and timing metadata for long-running processes.
- [DATA_EXFILTRATION]: The skill reads and writes job metadata (command arguments, timestamps, and exit codes) to local directories. This information is used exclusively for job status tracking and session recovery, and no unauthorized network transmission was found.
Audit Metadata