The Agent Skills Directory

[SAFE]: The skill implements a robust, multi-stage conversion process (AC to JSON Plan, then JSON Plan to Playwright Spec). This design limits the agent's autonomy during code generation, reducing the risk of malicious or unintended code output.\n- [SAFE]: Intermediate JSON test plans are strictly validated using a schema defined with the zod library. This ensures structural integrity and prevents malformed data from reaching the code translation layer.\n- [EXTERNAL_DOWNLOADS]: The project uses standard Node.js dependencies (zod, typescript, vitest) from the official npm registry. No suspicious external scripts or remote code execution patterns were found.\n- [COMMAND_EXECUTION]: The skill uses local CLI tools (npx generate-tests, npx validate-plan) for its core functionality. These tools are part of the skill's own package and operate on local files provided by the user.\n- [DATA_EXFILTRATION]: No network exfiltration patterns, hardcoded credentials, or unauthorized access to sensitive files were identified. File operations are limited to reading input AC/Plan files and writing generated artifacts to user-specified directories.

accelint-ac-to-playwright