loop-ops
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The check-native-facts.py script performs network requests to code.claude.com to verify scheduling limits against official documentation. These operations are part of a documented synchronization process and target a trusted platform domain.
- [COMMAND_EXECUTION]: The skill provides scripts for preflight environment checks and task orchestration. Tools like loop-doctor.sh verify that required binaries are available on the system PATH before a scheduled task begins, ensuring execution reliability.
- [INDIRECT_PROMPT_INJECTION]: By triaging external data such as pull requests or CI logs, the skill manages an attack surface for indirect prompt injection. It mitigates this risk through a defined risk-tier ladder and instructions for using boundary markers and safety guards.
- [DYNAMIC_EXECUTION]: The loop-scaffold.sh utility generates local wrapper scripts for task execution from bundled templates. This process is transparent, local to the environment, and does not involve the execution of untrusted remote code.
Audit Metadata