The Agent Skills Directory

[SAFE]: The skill operates exclusively on local files provided by the user (typically in a 'dist' directory). It uses fs.readFile and fast-glob to process HTML and robots.txt files for analysis.
[SAFE]: There are no network operations, subprocess executions, or dynamic code evaluations (eval/exec) present in the source code. All analysis is performed via static string matching and DOM inspection using the jsdom library.
[SAFE]: Dependencies are limited to established, well-known libraries (jsdom, fast-glob, tsx, vitest). The project maintains a clean package.json with no suspicious or unversioned dependencies.
[SAFE]: The skill includes a dedicated security scanner (tests/security/scan-forbidden-eeat.ts) that checks its own source code for prohibited patterns, such as placeholder domains, hardcoded ratings, or incorrect schema recommendations, demonstrating a strong security posture.
[SAFE]: No obfuscation, data exfiltration patterns, or persistence mechanisms were found. The logic is transparent and focuses entirely on the stated purpose of auditing E-E-A-T signals.

eeat-signals