local-ci

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands such as pnpm run ci:local and pnpm exec turbo run to mirror CI jobs locally. This behavior is expected as part of its primary functionality.\n- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface.\n
  • Ingestion points: Repository source code and pull request details used to select jobs.\n
  • Boundary markers: Absent; there are no delimiters to separate untrusted repository content from instructions.\n
  • Capability inventory: Command execution via package manager scripts and Turbo (SKILL.md).\n
  • Sanitization: Absent; the skill does not validate or sanitize repository data before use in command construction.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 11:10 AM