run-gui-journeys
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes repository-local shell scripts (
scripts/test-macos-journeys.sh) and TypeScript files (e.g.,scripts/test-vm/contract-journey.ts) usingbash,pnpm, andtsx. These operations are used to manage the lifecycle and testing of macOS GUI journeys within Tart virtual machines. - [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface where it ingests and processes data from external sources and virtualized environments.
- Ingestion points: The skill reads guest artifacts, provisioning logs, and 'native smoke resources' from paths like
~/.local/state/opensteer/vm-runs/and~/.cache/opensteer/vm/. - Boundary markers: No explicit boundary markers or delimiters for untrusted data are defined in the instructions.
- Capability inventory: The skill has the capability to execute shell commands, manage VM lifecycle via Tart, and run scripts via
pnpmandtsx. - Sanitization: There is no evidence of sanitization or validation of the artifacts or resource payloads before processing in the provided instructions.
Audit Metadata