validating-pre-commit
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Executes standard development commands via pnpm and npx for building, testing, and linting. These are restricted to the project local toolchain.
- [DATA_EXFILTRATION]: References environment variables ($SALEOR_URL, $SALEOR_TOKEN) for local E2E testing. This follows standard security practices for managing sensitive credentials and utilizes vendor-specific infrastructure.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and acts upon output from linting, build, and test tools. (1) Ingestion points: tool outputs from tests, build, and linting (SKILL.md). (2) Boundary markers: None present. (3) Capability inventory: Bash access for pnpm, npx, and rm (SKILL.md). (4) Sanitization: None present.
Audit Metadata