agentic-lookml
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze SQL queries provided by the user via the
lkr parse sql-lookml-comparetool to perform refactoring and modeling. - Ingestion points: The agent processes external SQL queries from the user which are deconstructed into LookML (SKILL.md, references/).
- Boundary markers: The instructions do not specify explicit delimiters to isolate the untrusted SQL input from the agent's instructions.
- Capability inventory: The skill uses CLI tools for file manipulation, project validation, and Looker API interactions.
- Sanitization: No specific sanitization or escaping of the SQL content is described.
- [EXTERNAL_DOWNLOADS]: The skill references and utilizes LookML writing guides from the
looker-open-source/looker-skillsGitHub repository, which is an official resource from a well-known service provider. - [COMMAND_EXECUTION]: The skill directs the agent to use the
lkrdeveloper CLI tool (a vendor resource) to parse SQL, generate LookML views, validate project schemas, and manage deployment to development environments.
Audit Metadata