data-lineage-mapper

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from multiple sources to generate lineage maps.
  • Ingestion points: The skill reads SQL query strings, dbt manifest files, Airflow DAG scripts, and Spark job definitions as input via the 'sources.content' parameter.
  • Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following instructions embedded in the processed code or metadata.
  • Capability inventory: The skill has access to 'Bash', 'WebFetch', and 'Read' tools, which allow it to interact with the file system and perform network operations.
  • Sanitization: No explicit sanitization or validation of the input SQL or manifest content is documented.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes external tooling for dbt project integration.
  • Evidence: The skill executes 'npx -y @dbt-labs/dbt-mcp' to facilitate manifest access.
  • Source: The '@dbt-labs' scope is the official repository for dbt Labs, a well-known service in the data engineering industry.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 07:58 AM
Security Audit — agent-trust-hub — data-lineage-mapper