Reputation Audit

Scores your own posts against a risk rubric with one LLM call per post, then rolls the verdicts into a 0-100 reputation score, a letter grade, and the specific posts worth a second look. Three surfaces share one scoring engine (src/ai/reputationScorer.js): a browser script that also renders a downloadable, shareable score card image; a CLI command for scripting; and an API route for agents.

This audits your own content for risk, not someone else's account and not follower quality. For follower/network audits, use Community Health Monitoring. For bulk-deleting by age, keyword, or engagement rather than AI-judged risk, use Content Cleanup.

The rubric

Every post is scored 0 (no risk) to 100 (severe) on up to four dimensions in a single model call, plus an optional custom question: