gcp-data-pipelines
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill identifies existing pipelines by scanning workspace files and their contents (e.g.,
dbt_project.yml,workflow_settings.yaml, or code imports likeorg.apache.beam) to determine the appropriate routing.\n - Ingestion points: Automated workspace scanning for framework indicator files and content strings as described in Step 1 of the skill.\n
- Boundary markers: Absent; the skill relies on file presence and internal string matching to branch logic without explicit input delimiters.\n
- Capability inventory: The skill acts as a router to specialized sub-skills (e.g.,
dbt-bigquery,gcp-dataflow) and provides guidance for direct execution using tools likegcloud.\n - Sanitization: None; the identification is based on static pattern matching within the development environment.
Audit Metadata