Phase 2: the loop and the model gateway. The composer in the console has
said "Not connected yet" since it was built, because there was no
assistant endpoint anywhere. There is one now.
- utils/chat.go the gateway, a sibling of embedding.go: one small
interface, a provider switch, the shared postJSON, no
framework. Agents name a TIER (fast/balanced/deep) and
config maps tier to model, so changing provider does not
touch an agent.
- services/assistantService.go one loop for every agent. An agent is a
name, a tier, a prompt and an allow-list — data, not a
class — so a sixth is config rather than a subclass.
- the endpoint under /v1/web, inheriting middleware.WebAuth along with
every other console route. The assistant reads the same data the console
does and must read it as the same person.
What the model does not get to decide:
whose data the caller is built from the verified session in the
controller, never from the request body — there is no
tenant field to fill in. A test scripts the model calling
a tool with {"tenantid": 916} and asserts it ran for 1147.
which tools the registry enforces the agent's allow-list; a test
scripts a call to a tool the agent lacks and asserts the
handler never ran.
when to stop steps and tool calls are counted here. A model that keeps
calling tools is stopped by arithmetic, not by being
asked nicely.
Two quiet failures have tests of their own. A finish_reason of "length"
means the provider cut the reply off mid-sentence, which reads exactly
like a complete answer unless it is flagged. And a truncated tool result
reaches the model in words it will repeat — otherwise it describes a
capped list and an empty one identically.
A refused tool goes back as a message, not an error: a model told "that
tool needs a tenant" can explain it, where a model handed nothing says
"something went wrong".
Optional, like the embedder. Without ASSISTANT_PROVIDER the endpoint
answers "not switched on here", the composer stays disabled, and the tools
still work — they are ordinary Go functions, and only turning a sentence
into a tool call needs a model.
14 tests, against a scripted model rather than a live provider: these are
about what the loop refuses to let a model do, and that has to hold for
any model, including one behaving badly.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The /web endpoints have never had authentication. The console keeps its
login record in per-tab sessionStorage and sends no Authorization header,
so every endpoint under /v1/web read `tenantid` off the query string and
believed it — one number in a URL reached another merchant's orders,
stock, staff and takings. `createposuser` under /v1/web/tenants minted
till credentials on the strength of an unauthenticated request, which the
route file already flagged in as many words.
Closed the same way posauth.go closed it for the terminals, in the same
order: the caller holds a token this server signed, and the tenant they
name is the tenant inside that token.
- utils/webtoken.go same HMAC construction as the POS token, 12h TTL,
a `w1.` prefix so the two kinds cannot verify as
each other
- middleware/webauth.go verifies the token, pins the tenant, and checks
a named branch belongs to it; reads the tenant from
the query, the body, and inside a JSON array, since
createdeliveries posts one
- login now issues the token; the console sends it as Bearer
Platform access rides on issuperadmin and nothing else. Not the role —
app_roles calls roleid 1 "Super admin" and tenant onboarding wrote 1 for
every shop owner, so a role test would promote every merchant on the
platform. Not a zero tenant either, or a user row with the field unset
becomes the one session that reads everything. Both near-misses have
tests.
WEB_AUTH_REQUIRED defaults to off. The console in production does not
send a token yet, and enforcing before it does would lock every merchant
out of a working product. A token that IS sent is always verified, and
one naming the wrong tenant is always refused; the flag only decides what
happens to a request carrying none. This should be a short-lived state.
Still trusting the caller: partnerid, customerid and appuserid, which
some list endpoints also scope on. Noted in the middleware header.
25 tests.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
POST /v1/mob/scan/lookup label + customer → catalogue match, sizes, and
every registered store that sells it with live
stock, in-stock first / nearest first, one
recommended
POST /v1/mob/scan/confirm chosen store + size + qty → re-read the ledger;
ok, or the next-nearest store with enough of the
same product
GET /v1/mob/scan/stores registered stores nearest first
Recognition is pgvector cosine search over every brand_* table (each
with its own index, merged) plus a word match that settles near-ties
and works alone when no model is configured. The embedder is chosen by
EMBEDDING_PROVIDER (OpenAI-compatible or Gemini) and must be the model
that indexed the catalogue: verified 2026-09-15 as all-MiniLM-L6-v2 over
search_query, served by the cluster's Ollama as `all-minilm`; the first
search refuses a width mismatch by name.
Customer, stores and catalogue are read concurrently under a 5 s cap; a
slow model degrades to a text answer. Vectors and ranked hits are cached
in Redis and in-process; live stock never is. Availability uses the same
rules as the customer catalogue (approve, publishedat, ledger balance,
outlet price else retail). No stock reservation: confirm re-reads.
scratch/cataloguedims reports the catalogue's embedding width and fill.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
A till holds every bill in its own SQLite database and keeps it for
seven days after we acknowledge it, marking one synced only when its id
comes back in an ack. Everything here follows from that.
Silence is not acceptance, so a failing ingest publishes nothing at all
and the terminal simply sends again. A duplicate is a success, because
at-least-once delivery means a lost ack legitimately re-delivers bills
we already hold, and calling those failures would strand a day of
takings on the till. Deduplication is a unique index on the terminal's
UUID plus an advisory lock held for the transaction.
Bills land in pos_orders / pos_order_items rather than orders: a counter
bill carries a cashier, a terminal, a rounding adjustment, promos,
loyalty movement and a payment split that orders has nowhere to put, and
forcing one into the other loses whatever does not fit. Stock is *not*
split — a counter sale writes the same productstocks rows an app order
does, through helpers extracted from createOrderTx so the rule that
prevents overselling has one implementation rather than two.
GetRevenueSummary and GetSalesSummary were extended to union the new
table in; any new report has to remember the same.
Terminal health goes to Redis under a 90-second TTL, sharing the
instance the express backend uses. A heartbeat is a fact with an expiry
date: a till that loses power stops refreshing and ages off the board by
itself, where a Postgres row would need ~288k writes a day and a reaper.
Proven end to end against the live estate before commit: a bill over
HTTP and one over the real Mosquitto broker, the same bill three times
producing one row and one stock movement, and a heartbeat arriving on
the health endpoint. All probe data was removed afterwards.
Four things that only surfaced against real data. An unset jsonb column
failed the very first bill. Product SKUs are unusable as barcodes — 6,245
products share 93 SKUs and "1" covers 5,794 of them — against the till's
unique index, so barcodes fall back to the product id. A taxpercent of
-1 exists and would have put negative GST in a filed slab. And a product
with id 0 exists, which can never be billed and is now skipped.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Adds a separate CatalogueDB (pgvector) connection alongside the main
nearledb, plus a new catalogue module (repository/service/controller/
routes) to browse it by brand, category, and keyword, with brand
optional so the whole ~237-product catalogue can be browsed unfiltered.
Adds the actual bridge: importing a catalogue product snapshots it into
the tenant's own products table (keyed on brand+catalogueid, since a
catalogue row's bare id is only unique within its own brand table),
then links it via the existing productlocations upsert. Re-importing
tops up stock and refreshes price instead of duplicating. Also adds an
imported-refs endpoint so the frontend can badge already-imported items
without diffing full product lists, and wires the new AWS S3 image
store used to resolve catalogue product photos.
Bumps Go/Docker to 1.24 for the AWS SDK dependency this needs.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
