turbo-transforms

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documents the use of the fetch_abi (also aliased as _gs_fetch_abi) SQL function, which retrieves contract ABI or IDL definitions from remote URLs. This is a standard feature of the Goldsky platform used to facilitate the decoding of blockchain event logs and does not represent an unauthorized download.
  • [COMMAND_EXECUTION]: The agent is instructed to use vendor-specific CLI commands, such as goldsky turbo validate and goldsky turbo inspect, for verifying pipeline configurations and debugging transform outputs. These commands are part of the official Goldsky development environment.
  • [DATA_EXFILTRATION]: The skill demonstrates safe secret management practices by using placeholders like secret_name: MY_SECRET in sink configurations, rather than hardcoding sensitive credentials. No patterns of unauthorized data transmission were found.
  • [PROMPT_INJECTION]: As the skill involves the agent processing and explaining user-provided pipeline YAML and SQL files, it possesses an indirect prompt injection surface. However, the skill's logic is constrained to technical assistance and the use of validation tools, which is appropriate for its stated purpose as a developer utility.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 09:02 PM
Security Audit — agent-trust-hub — turbo-transforms