verifier-setup
Pass
Audited by Gen Agent Trust Hub on Jul 21, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface by reading configuration data from repository files (such as package.json and Makefile) and interpolating those values into a new skill file. If these source files contain malicious instructions, they could be propagated into the generated verification workflow.\n
- Ingestion points: Repository configuration files identified in Step 1.\n
- Boundary markers: The template file uses markdown headers but lacks specific instructions to ignore embedded commands in variables.\n
- Capability inventory: The resulting skill is capable of executing shell commands for stack orchestration and testing.\n
- Sanitization: The process does not explicitly validate or sanitize strings extracted from the repo before interpolation.\n- [EXTERNAL_DOWNLOADS]: The skill uses npx to ensure the @playwright/cli package is installed. This is a well-known and standard package for browser automation tasks.\n- [COMMAND_EXECUTION]: The skill executes shell commands to inspect the repository, install the Playwright driver, and create new skill artifacts. It also triggers external setup skills for local development and sandboxing.
Audit Metadata