Inject

Any data item

Connectors or approved one-off imports load email addresses, cookies, events, CRM rows, products, support, billing, documents, and outcomes.

Store

Attribution paths

Scout stores each item's relationship set, vector, and ordered attribution path in customer-owned PostgreSQL/pgvector. Fortress moves high-load memory into Rust/LanceDB, not KynticAI product operations.

Compare

Rust + LanceDB

Enterprise contains the proprietary Rust relationship, weighting, traversal, and LanceDB path for fast similarity analysis across millions of relationship sets.

Output

Top-example JSON

The engine creates JSON with the best examples, importance bands, confidence, caveats, and ranked task options for the question being asked.

Explain

LLM task brief

Fortress sends JSON to your approved model boundary. The proposed Elite path shows the executive walkthrough from safe discovery to scoped pilot and outcome review.

Review Product Proof

Architecture / KynticAI Agentic Importance Framework

The local-first agentic layer over evidence-weighted importance.

KynticAI Agentic Importance Framework turns importance scoring into a private decision-support workflow: gather evidence, score what matters, identify missing data and contradictions, build compact model context, route to local loopback Ollama, an internal gateway, an OpenAI-compatible gateway, or an Anthropic-compatible gateway under deployment policy, then generate a safe answer, next action, or dossier.

View KynticAI Agentic Importance FrameworkPlatform Architecture

Core layers

How KynticAI Agentic Importance Framework is structured

01

Evidence gatherer

Collect approved facts, summaries, source references, prior outcomes, and review notes from the local workflow boundary.

02

Importance scorer

Rank evidence by outcome relevance before the agent asks questions, routes work, or spends model tokens.

03

Question and contradiction layer

Identify missing variables, conflicting evidence, stale signals, and review needs before generation.

04

Context builder

Prepare compact model context with high-signal evidence, compressed secondary evidence, deferred low-signal material, provenance, and caveats.

05

Runtime router

Route to local loopback Ollama, an internal model gateway, an OpenAI-compatible gateway, or an Anthropic-compatible gateway under deployment policy.

06

Dossier reporter

Generate safe answers, next actions, forensic dossiers, missing-data notes, and provenance-aware reports.

Operating flow

The request path through the product

Gather

Gather evidence

Start from approved local evidence rather than asking an agent to scrape or improvise context.

Score

Score importance

Promote the facts that matter most and keep weaker evidence available without forcing it into every model call.

Ask

Ask for what is missing

Turn missing data and contradictions into better questions before the answer is generated.

Route

Route through the approved model path

Use loopback Ollama, internal gateways, OpenAI-compatible adapters, or Anthropic-compatible adapters under deployment policy.

Report

Report with provenance

Produce decision-support output with evidence references, caveats, missing-data notes, and review boundaries.

Monitor

Monitor the review state

Track caveats, review needs, and outcome feedback without turning product operations into the customer data plane.

Example signal path

A churn-risk agent produces a safe, provenance-aware dossier

Illustrative sample only. Deployment policy decides whether the model route is local Ollama, an internal gateway, an OpenAI-compatible adapter, or an Anthropic-compatible adapter.

01 / Source

Example fields

workflow = churn_risk_review

sources = CRM + support + usage

runtime = loopback_ollama

scope = customer_controlled

02 / Evidence

What KynticAI creates

importance_band = elevated

missing_data = renewal_owner

contradiction = usage_up + complaint_open

deferred = low_signal_notes

03 / Action

What the business does

generate next_action_brief

include missing_data_notes

retain provenance and caveats

route to human review

Operating model

How the product stays useful at enterprise scale

Weights

Protected scoring values

Outputs use reason categories, importance bands, and caveats while private scoring values stay inside the scoring boundary.

Tables

Private scoring tables stay protected

Generated dossiers receive context plans and explanation bands while proprietary scoring tables remain inside the protected runtime.

Runtime

Customer-controlled conversation data

Deployment policy decides whether work routes locally, through an internal gateway, or through an approved hosted-provider adapter.

Adapters

Approved model routes

Local Ollama, internal model gateways, OpenAI-compatible gateways, and Anthropic-compatible gateways are governed deployment options.

Integration points

Where it connects to the wider stack

Ollama

Loopback Ollama

Local model route for private agent proof and customer-controlled runtime paths.

Adapters

Approved provider adapters

Hosted providers such as OpenAI and Anthropic can be reached through approved OpenAI-compatible or Anthropic-compatible deployment adapters.

Workflow

Enterprise decision surfaces

Connect to internal review queues, sales/churn workflows, credit and hiring review, operations, supply chain, diligence, and support.

Dossiers

Safe dossier outputs

Return next actions, explanations, forensic dossiers, missing-data notes, and provenance-aware reports.

Evidence Results

Agentic Importance turns scored evidence into safe local-first dossiers.

These examples show the agentic path: gather evidence, score importance, ask for missing data, route to loopback Ollama today, and report with provenance.

KynticAI Result
Sales Agent

Agentic Importance scenario - CRM, call notes, product usage, missing procurement data

Ask the question that changes the deal

“The agent gathers deal notes, product usage, objection history, and support signals, then marks procurement owner as missing before drafting a next-action dossier.”
KynticAI Result
Operations Agent

Agentic Importance scenario - supply chain incident, tickets, shipment events

Turn operational noise into a reviewable dossier

“Supplier delay, warehouse backlog, and customer-impact notes are ranked before the agent generates an incident dossier.”
KynticAI Result
Private Runtime

Agentic Importance scenario - loopback Ollama, approved hosted adapters, customer-controlled routing

Give buyers deployment choice

“The model route can use local loopback Ollama, an internal model gateway, an OpenAI-compatible adapter, or an Anthropic-compatible adapter under deployment policy.”