Settings

Control scan safety, model use, and workspace defaults.

This page is scaffolded for later credentials, scheduling, and cost controls. Phase 2 keeps implementation local and testable.

Policy

safe-scan-policy-v1

API-first, no aggressive scraping

Live scans

disabled

Env-gated before credentials or writes

Scan cap

25 items

4 sources per run

Dry-run cost

No live LLM calls required

Workspace defaults

Workspace

Scan posture

No credentials configured

Reddit and other gated sources will stay disabled until explicit credentials and safe scan policy are added.