Grounded truth
Every answer points to a real stored fact — traceable to the row it came from. If it isn't there, FaultLine says so. It never bluffs.
Saves you tokens — two ways
It returns the one true fact, not 40k tokens of history every turn — smaller prompts. And your agent keeps its own memory of what worked and what failed, so it stops re-solving solved problems and re-hitting dead ends. Fewer wasted calls, lower spend.
Teach it a whole field
/expand learns an entire domain — networking, law, medicine — once. Everything you say after lands in the right place, and is instantly walkable.
One seat, three prices — the monthly base is your tenant portal AND it buys every seat down. Individual (no base, first month free): Standard $15 · Advanced $40 · Expert $60 · Agent $120/mo. Team — everything in Individual, plus the tenant portal, on a $750/mo base (incl. 2 Standard seats, up to 15) that drops seats to $12.50 / $35 / $50 / $100: the affordable step up before enterprise. Enterprise — everything in Team, plus unlimited seats, SSO, dedicated isolation, SLA & support on a $1,500/mo base (incl. 5 Standard) with the lowest seats $10 / $30 / $45 / $80. The same seat costs less the more base you commit. See every price ▸
Building on it? There's a developer API (V1, in preview) — the same validated memory over a /v1 REST endpoint and a Python SDK (pip install faultline), so you can put grounded memory in your own app or agents. See the API on the geek page ▸