skills/gemini-cli-extensions/data-agent-kit-starter-pack/gcp-managed-airflow-migrations/Gen Agent Trust Hub
gcp-managed-airflow-migrations
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses local shell commands, including
grepfor pattern discovery andrufffor static analysis of Python DAG files to identify deprecated syntax or breaking changes. - [REMOTE_CODE_EXECUTION]: The skill utilizes the
gcloud composer environments runcommand, which allows for the execution of Airflow CLI commands (likedags listanddags list-import-errors) within the remote managed environment's worker nodes. - [EXTERNAL_DOWNLOADS]: The skill facilitates the transfer of DAG files and environment configuration data from Google Cloud Storage buckets to the local workspace using the
gcloud storage cputility. - [PROMPT_INJECTION]: The skill processes external DAG files which could theoretically contain malicious instructions designed to influence the agent's behavior during migration.
- Ingestion points: Processes DAG files downloaded from GCS buckets or located in the
./migration_workspacedirectory. - Boundary markers: The instructions do not define explicit boundary markers or warnings to disregard instructions embedded within the files being scanned.
- Capability inventory: The agent uses
grep,ruff, andgcloudtools, which possess significant capabilities for file system access and remote environment interaction. - Sanitization: There is no explicit sanitization or validation of the content within the DAG files or their filenames before they are passed to shell command templates.
Audit Metadata