With DISPATCH_AGENT autonomous=true in the agent registry, the assignment-failure
path sends a real customer message. It would have sent
"Delay alert: Order 36 may arrive later than expected. New ETA: {new_eta}"
with customer_id "unknown": the DELAYED template names {new_eta} but DispatchAgent
supplied only order_id, and it has no customer id to give.
- _send_via_channel refuses to send when any {placeholder} survives substitution
(or the template is missing), logs which variables the caller omitted, and
records a failed notification. Silence beats a broken message.
- customer_id is omitted from the payload when it is absent or "unknown" rather
than sent literally; the backend resolves the recipient from order_id.
- DispatchAgent supplies new_eta "being confirmed" — honest, since no ETA exists
at assignment-failure time.
- Tests: the old bug is refused, a complete payload renders clean, the id is
omitted/passed correctly, and DispatchAgent's vars satisfy the template.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012AJLYcbTHCe45fyFnMfEin
Every booking.assignment_failed observed in production carries no zone_id, so
the per-zone failure counter and the once-a-day alert limit collapsed into a
single global "unknown" bucket — one alert per day for the whole country.
zone_key() prefers the backend's zone_id and falls back to a coarse lat/lon
grid cell (1 dp, ~11 km, DISPATCH_GEO_BUCKET_DP). The failure reason is now
logged too.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012AJLYcbTHCe45fyFnMfEin
The previous commit treated a missing miler_status:<id> key as "not available".
In production only 2 of 34 milers have that key at all, so the agent would have
reported "no available rider" for nearly every failure — a confident wrong
answer in the opposite direction from the bug it fixed.
- _miler_presence returns available / unavailable / unknown. No key, an
unparseable value, or an unrecognised status reads as unknown.
- _find_zone prefers a confirmed-available miler, otherwise reports the nearest
unknown-presence one (it may well be assignable), and returns None only when
every nearby candidate is confirmed off duty.
- Facts carry nearest_miler_within_km + nearest_miler_presence; the prompt
states plainly that unknown presence is not evidence of a coverage gap and
should lean to monitor/escalate rather than ops_alert.
- New eval case for riders-nearby-but-no-presence-data; tests for all three
presence states.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012AJLYcbTHCe45fyFnMfEin
Prod findings (2026-09-22): a backend retry sweep failed 1,001 bookings in
60s; the agent called each a "coverage gap" because GEORADIUS found a miler
in the geo index (last seen in June), then sent 1,001 ops_alert tasks to
CUSTOMER_AGENT, which has no such handler.
- _find_zone filters GEORADIUS candidates by the backend's miler_status:<id>
key; only status=Available counts. Facts now carry
nearest_available_miler_within_km plus milers_in_geo_index_within_30km so
the decision can separate "no riders here" from "riders exist, none on duty".
- Rate limit per zone per day: after the first alert, further failures only
bump the counter (no LLM call); a summary re-alert goes out every
DISPATCH_REALERT_EVERY (default 100).
- _ops_alert / _escalate_dispatch send EXCEPTION_DETECTED to JARVIS (the path
that is actually handled); customer delay notice uses CUSTOMER_AGENT's real
send_notification contract.
- JARVIS: escalation inbox (_escalations, pending_escalations()) and
human_review/ops_alert task types are recorded instead of dropped.
- ExceptionAgent pull loops: also catch asyncio.TimeoutError (distinct from
nats.errors.TimeoutError on 3.11) and log the exception type — the blank
"pull loop error:" lines.
- Prompt + eval cases updated for the renamed facts; new case for the
observed index-full/nobody-on-duty pattern. Tests for liveness filtering,
burst suppression, fallback heuristic, sinks, and the JARVIS inbox.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012AJLYcbTHCe45fyFnMfEin
New agent that owns the DoormileExpress dispatch flow. The normal B2C booking
flow is untouched; this handles express orders that a console operator has let
accumulate batch by batch, then dispatches in one go.
Triggered by express.dispatch_requested (the console's manual dispatch action).
For the tenant it: reads the pending bookings and the tenant's available riders
(Go internal API), distributes them across riders by proximity + load with a
radius guard, asks routes.workolik.com (Valhalla) for each rider's stop order,
and writes the assignments back through the Go internal API — which stays the
single writer of assignment state; the agent only decides who and what order.
- agents/express_dispatch_agent.py: the agent (mirrors DispatchAgent's NATS
binding; pure consumer of the backend-owned EXPRESS stream).
- config/system_config.py: ROUTE_OPTIMIZER_URL.
- registered under JARVIS in main.py / agents package.
- tests: distribution (radius, load, cap, no-GPS fallback) and sequencing
(single-stop, remap, optimizer-down, unknown-id) — deterministic, no network.
Gated off by default on the backend (EXPRESS_AGENT_ENABLED); EXPRESS_AGENT_
AUTONOMOUS=false runs it observe-only (logs the plan, writes nothing).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
