proofreader-api
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed to assist developers in implementing the experimental browser Proofreader API. It follows standard integration patterns and provides helpful templates and documentation.
- [COMMAND_EXECUTION]: The skill includes a Node.js script (
scripts/find-proofreader-targets.mjs) used to inventory the workspace for frontend files and existing API markers. This script operates exclusively on the local file system to search for relevant strings and does not exhibit any network activity or unauthorized command execution. - [DATA_EXPOSURE]: Analysis of the discovery script and code templates shows that data processing is restricted to identifying API-related strings within the project workspace. No sensitive data access or exfiltration patterns were detected.
- [PROMPT_INJECTION]: The instructions focus purely on technical implementation and debugging of the Proofreader API, with no attempts to override system prompts or bypass safety filters.
Audit Metadata