The problem
Then the work starts on revenue numbers, not profit. Most clients have never calculated break-even ROAS; Northline had, and the agent still checked it.
The insight
get_next_item submit_answer get_progress. 32 checklist items, one question per turn.How it works
Asks, groups, drafts, splits, pre-checks, summarises. Rules first, so it runs with the model down; the local model adds polish. It has no approve, send or delete tool.
Approves the spec G1 · accepts tickets G2 · reviews, never the builder G3 · releases, with client sign-off G4 · sends G5. Open gaps flow back to the client's AI through intake.
The money moment · SPEC-003 Profit model · Northline Goods
Floors are derived by the agent from the client's own R2 answer and marked ≈; the client's R3 break-even adds shipping and returns. A number that traces to no answer and isn't marked ≈ fails the review pre-check.
Guardrails
Local-first by construction
:8787/mcp:8788:8790:8792:8791/mcpevery 3 sloopback only · 20 s timeout · rules fallbackLive demo · 60 seconds · console at 127.0.0.1:8792
Why it's a business
up to 20 active clients; the box bought or leased.
12 h × $100/h ≈ $ saved per onboarding. Five a year cover it. Fixed box, not a token bill.
Why now: owners already work inside AI assistants, and MCP lets them call another company's tools · a desk-side 128 GB box runs capable open models · NVIDIA's sandbox makes an unattended agent acceptable.
Built today · what's next · the ask