Injection-safe by design
Facts the model reads from a chat, an image or a retrieved document never auto-write to your profile; they queue as confirm/reject suggestions, so a hidden instruction can't quietly poison every future answer.
Harvora Intelligence™
A decision-intelligence layer, built for horticulture
23 AI features, and every one exists because of a Harvora prediction. The findings are deterministic; the AI puts them into words and reasons over them. Grounded, auditable, and always yours to confirm.
Grounding before generation
No AI output touches a number it didn't retrieve
Every AI-generated report, briefing, insight and answer runs the same five-step discipline. Hallucinated figures aren't rare; they're architecturally impossible.
Grounding pipeline · every answer01 → 05 · every block · every day
- Step 01
Query
Every relevant prediction, observation and dollar figure (including actual costs available when a read-only Xero connection is authorised and freshly synced) is retrieved from your blocks' database first.
- Step 02
Bundle
Assembled into a structured context bundle. The AI only ever sees verified figures.
- Step 03
Narrate
The model synthesises and narrates. It never computes a number on its own.
- Step 04
Validate
Output is parsed against a strict schema before a grower ever sees it.
- Step 05
Fall back
If validation fails, a deterministic structured display ships instead. No malformed AI output, ever.
While you sleep
Five detectors sweep every block, every night
At 2:30 am the nightly scan runs deterministic detectors over each block's prediction history. Only genuine findings surface, deduplicated against alerts you've already seen, narrated in plain language by morning. The AI Season Planner reschedules frost, irrigation, harvest and scouting events on every forecast update, your 5 am Daily Briefing ranks the day's decisions by expected value, and a pre-decision checklist spells out what to verify before you act.
- 01Trajectory changeThe forecast midpoint has shifted beyond the block's own rolling variability.
- 02Cross-prediction correlationAgronomist-reviewed rules across surfaces, e.g. sustained disease risk while yield softens.
- 03Historical anomalyToday's value sits far outside the same block at the same point in past seasons.
- 04Accumulating riskCold hours, disease-favourable days or water deficit building without any single-day alert firing.
- 05Unexplained deviationA prediction lands outside the confidence band the block's own season model expects.
Scan02:30Briefing05:00
The showpiece
Ask Harvora™: an agent that knows your growing operation
Ask anything about your operation and get an answer grounded in your blocks, your history, your numbers, right down to actual costs available when a read-only Xero connection is authorised and freshly synced. Under the hood it's a bounded agentic loop, powerful enough to run what-if scenarios, disciplined enough to never act without you.
Agent onlineAsk Harvora
Grounded inYour blocksYour historyYour numbers
Drafts, answers, proposes. Never acts without you.
01
Current evidence
Every critical answer carries typed prediction identity, model version and a validity window. Related calendar evidence is reconciled as current, superseded, expired or unresolved before the agent responds.
02
Agentic tool-calling
A bounded agent loop over read tools, a scenario runner and tiered action tools: hard budgets, one repair retry, deterministic fallback.
03
Grounded retrieval (RAG)
Hybrid full-text + vector search over your own records, rank-fused, isolated to your orchard in SQL. Citations shown.
04
Self-consistency
High-stakes questions are answered three times independently and you get the consensus; if the runs disagree, that disagreement is flagged to you rather than hidden.
05
Voice
Hands-free in the orchard: questions spoken, answers streamed back sentence by sentence.
06
Partner MCP
External agents can read and propose through a consent-gated endpoint. They can never execute a write.
AI alert digests
The urgent calls, delivered once, while there is time to act
Critical and watch findings are bundled into one coordinated SMS and email digest. Harvora suppresses repeats, carries escalation forward, respects quiet hours and only breaks the hold when waiting would close the action window.
- 01Analysis-backedEvery digest starts with a real critical or watch item from the same analysis layer that drives Today's Priority.
- 02One threat, one threadFingerprint deduplication prevents reworded repeats, while a watch escalating to critical supersedes the pending alert instead of double-paging.
- 03Quiet by defaultAlerts are bundled and rate-limited per recipient, with quiet hours holding delivery unless waiting would miss the action window.
- 04SMS + mirrored emailBoth channels carry the same digest at the same moment, from one delivery clock and one source of truth.
Quiet hoursrespectedDigestSMS + email
Season over season
Intelligence that compounds, block by block
Every season you grow with Harvora, four learned models deepen for each block. Each carries an explicit insufficient-data status; nothing is fabricated on a young block. What they learn flows back through the whole engine: forecast corrections, expectation bands, risk thresholds and briefings, all tuned to how each block actually behaves.
- 01Microclimate correctionsCompare block temperature and humidity observations with the regional forecast across clear nights, advection fog and nor'west flow. Harvora surfaces the pattern-specific difference so a reviewed local correction can strengthen the next decision.
- 02Disease pressure antecedentsTrace the weather conditions that precede elevated Psa pressure on each block. Harvora provides the regional disease-pressure signal now; block observations help the AI learn how aspect and shelter shape leaf wetness locally.
- 03Yield-weather responseRead how weather conditions map to each phenological stage in your block's yield outlook. Confirmed harvests make the response increasingly block-specific season over season.
- 04Irrigation responseTrack measured moisture evolution alongside rain and evapotranspiration. Connected soil observations let the reviewed response incorporate the drainage and soil properties the map cannot see.
Hardened for the field
A reliability layer under all 23 named features
An end-to-end adversarial audit hardened every reliability edge of the AI layer, the part that keeps it trustworthy under load, bad input and provider failure. Each item below shipped with tests, and the model-facing ones with a passing evaluation.
Honest when it can't finish
A truncated or malformed generation says so and degrades cleanly, instead of stopping mid-sentence or masquerading as an outage.
Consensus on the big calls
High-stakes answers are generated multiple times and the response you see is the one the runs agree on, judged by what the answers mean, not how they're worded.
Budgeted and bounded
A rolling per-orchard cost budget covers even the agent stream and block ask; context and tool results are capped, with an explicit “trimmed” note rather than a silent drop.
Truthful operational status
Provider retries preserve useful coverage, partial recovery stays visible as recovering, and optional unconfigured services are kept distinct from genuine degradation.
The AI drafts. You decide.
Harvora's AI suggests, schedules, and generates, but nothing writes to your operation without your explicit confirmation. Human judgment stays exactly where it belongs.