One AI-native foundation. Many sector products.

Every OpenLI product grows from the same foundation: OpenLI Codex — the OpenRunner agent-runtime fabric, with our own third-generation OpenCodex harness running alongside the Claude Agent SDK and OpenAI Codex SDK as interchangeable runners behind one protocol. Each sector branch then specialises for a regulated industry, inheriting the foundation’s governance, multi-tenancy and audit trail.

Our vision: in tomorrow’s software, natural language is the only programming language enterprises need to build, modernise and operate mission-critical systems — with better quality, traceability and governance than any code-only stack can offer.

Click any product card to open its page. GitHub icons link directly to source. Pipeline products are marked with an amber dashed border.

FOUNDATION TIER · ROOT OF THE PRODUCT FAMILY

OpenLI Codex

OpenLI Codex is the agentic AI foundation that powers every other OpenLI product. It is the OpenRunner agent-runtime fabric — one protocol, every runner, metered — with our own third-generation OpenCodex harness at its centre. It is the technical and commercial differentiator: a purpose-built foundation for production agentic workflows in regulated environments — not a wrapper around LangGraph, LangChain, or any single vendor’s SDK.

FOUNDATION

OpenLI Codex OpenRunner fabric · OpenCodex runner

Our own third-generation agent runtime. Multi-runner. LLM-agnostic. Governance-first. Metered & audit-ready.

OpenCodex · in-house runner

Claude Agent SDK

OpenAI Codex SDK

Our own third-generation runner. OpenCodex is a clean-room enterprise harness — explicit plan→act→observe→reflect loop, permission tiers, agentic search (no vector RAG) — not a wrapper over a vendor SDK.
Multi-runner behind one protocol. Swap OpenCodex, Claude or OpenAI Codex per task, workspace, tenant and compliance zone — with zero consumer code change. No vendor lock-in.
LLM-agnostic & metered. Any model via a central gateway (Anthropic, OpenAI, Bedrock, vLLM, self-hosted); every run teed into a per-tenant cost ledger, with $0-token CI by default.
Governance-first by design. 7-tier RBAC, human approval gates, audit trails on every agent action, PII sanitisation hooks, durable resumable runs.
Open-source core. AGPL-3.0 with a commercial dual licence for enterprise — auditable by your security team.

Open OpenLI Codex Portal View on GitHub Talk to the team

HEALTHCARE & OPERATIONS

From building integrations to running them in production.

The Healthcare cluster covers the full lifecycle of an NHS Trust integration estate: design and build interfaces, resolve patient identity across systems, migrate legacy engines to modern platforms, modernise legacy IRIS code in plain English, monitor and triage in production, and migrate the underlying patient data when systems are upgraded.

OpenLI HIE

v1.9.7 · production · whitelabel-ready

The first GenAI-native healthcare integration engine. Natural language is the development language for HL7, FHIR and IRIS.

HL7 v2.x · FHIR R4 · HTTP/REST · file · database
7-role RBAC with clinical safety guardrails
Hot reload, config snapshots, one-command rollback

Open Portal GitHub

OpenMPI

v0.1.7 · preview · NHS Acute first

Agentic-AI-assisted Master Patient Index. One trusted patient identity across every system in your trust — AI explains ambiguous matches in plain English; stewards always make the call.

7-layer matching pipeline with AI rationale at the borderline
Built-in integration: HL7 v2 · FHIR R4 · IHE PDQm/PIXm · CSV · NHS PDS
DCB0129 / 0160 evidence; SHA-512 audit chain; bias regression on 4 protected groups

Open Portal GitHub

OpenLI TIE Migration

v0.8.12 · production · whitelabel-ready

Trust Integration Engine migration. AI-governed migration of NHS Trust integration engines to modern platforms, in plain English.

5-stage migration lifecycle with phase-gate approvals
Plug-and-play runners (Mock, Codex, Claude)
Reference benchmark: a UK NHS Foundation Trust

Read product page

IRIS CoPilot

vision sample · bundled with HIE

Natural-language modernisation for InterSystems IRIS. The canonical proof of the OpenLI thesis: natural language is the only programming language enterprises need to modernise legacy systems.

Plain-English ObjectScript generation and refactoring
Lifecycle support for productions and business hosts
Same governance & audit framework as HIE

GitHub

OpenTriage

v0.4.6 · early production · whitelabel-ready

Agentic AI alert triage and root-cause analysis for NHS Trust integration estates. Real-time telemetry, evidence-backed findings, audit-logged.

30+ pre-defined AI use cases on cron and event triggers
Connectors for IRIS, Rhapsody, Mirth Connect
Alert lifecycle with audit-logged transitions

Read product page

OpenLI Data Migration

v0.2.5 · production · whitelabel-ready

Enterprise data migration for NHS PAS & EPR upgrades. Contract-driven ETL with quality gates, semantic mapping and immutable audit trails.

Schema extraction & mapping (417 source → 38 target tables)
Quality gates and release gates per environment
Crosswalks, mock data, audit explorer

Read product page

PHARMA & CLINICAL

AI-native risk-based monitoring for the modern clinical trial.

Clinical trial risk-based monitoring is shifting from spreadsheets to real-time risk surfaces. The Pharma cluster brings the OpenLI governance-first agentic runtime to sponsors, CROs and biotech operators — auditable, explainable, and aligned with ICH-GCP and 21 CFR Part 11 expectations.

OpenTrials

v0.3.17 · live demo · ct.openli.ai

AI-native RBQM for the modern clinical trial. Risk-Based Quality Management, central statistical monitoring, signal review and FDA / EMA submission support. Built on the OpenLI Codex agentic AI foundation.

RBQM lifecycle: KRI definition → central monitoring → signal review → disposition
CDISC SDTM 3.3 / ADaM 1.3 / Define-XML 2.1 study-package import (folder, ZIP, or server-folder)
21 CFR Part 11 audit trail (SHA-512 hash chain), ICH E6(R2/R3) GCP, EU Annex 11, GAMP 5
5 demo studies preloaded (PHOENIX-3 + 4 Vanguard packages); 8 role-based logins

Open the demo GitHub

FINANCE & COMPLIANCE

Tax filing today. Personalised investment guidance tomorrow.

Finance & Compliance brings the OpenLI foundation to regulated finance buyers: UK SMEs filing Corporation Tax today, and banks & wealth managers offering AI-augmented investment guidance to their customers tomorrow — all with the same governance, RBAC and audit trail that healthcare and integration buyers rely on.

OpenCT

v1.1.4 · production · whitelabel-ready

UK Corporation Tax filing. CT600 v3 XML, iXBRL statutory accounts, and direct submission to HMRC. AI-native, open-source, and a fraction of the cost of commercial software.

FRS 105 & FRS 102 Section 1A iXBRL statutory accounts
Sage / Xero / QuickBooks / CSV import with auto-classification
HMRC Transaction Engine direct submission

Open OpenCT GitHub

OpenInvestor

Pipeline · Early Access

Coming soon · early access opens 2026 H2

AI-native investment guidance for banks & wealth managers. Personalised, explainable, and built for FCA Consumer Duty and MiFID II alignment.

Will support: explainable investment recommendations with full audit trail
Will support: suitability and appropriateness checks (MiFID II)
Designed for retail bank, private bank, IFA whitelabel deployment

Join the early access list

TRAVEL & EXPERIENCE

Journey-led leisure platforms beyond the experience economy.

The Travel & Experience cluster proves that the OpenLI foundation works beyond regulated B2B sectors. Journey-centric platforms, community contribution, partner attribution and B2B2C operations — all with the same governance and AI-native engineering that powers our healthcare and finance products.

GSJ Platform

v0.6.0 · production · Partner-led

AI-native, journey-centric enterprise platform for Great Scenic Journeys. B2C journey discovery, B2B partner attribution, and B2B2C community operations on one foundation.

Journey lifecycle (draft → review → approve → publish → sync)
Partner campaigns & attribution reporting
WordPress integration surfaces & rewards foundations

Open Portal GitHub

ENTERPRISE OPS

AI copilots for the SOPs, POs and bids that run your enterprise.

The Enterprise Ops cluster brings the OpenLI foundation to knowledge-heavy operations — from the Standard Operating Procedures and purchase-order exceptions that run merch and supply chain, to the proposals and bids that win professional-services work. Two sibling products on the same foundation: OpenSOP collapses the planner-reads-SOP-decides loop into seconds; and OpenBid (forked from OpenSOP) works a whole bid end-to-end, grounded in the firm’s own knowledge — both with full audit trace and a human firmly in the loop.

OpenSOP

v0.1.9 · live demo · sop.openli.ai

AI-native copilot for enterprise SOPs & POs. Triple-runner architecture (OpenLi Codex, Claude Agent SDK, dedicated pgvector RAG agent) emits one structured recommendation under three production guardrails. Reference deployment: ASOS Merch PO Exception Triage.

3 interchangeable runners on one recommendation contract · per-call toggle
pgvector HNSW cosine RAG over section-anchored SOP chunks · top_k=10
3 guardrails: PII scrub (in + out) · gated threshold-contradiction · low-retrieval-confidence
32/32 pytest · 20/20 Playwright in Docker · 9/9 evals across 3 runners

Open the demo Product page GitHub

OpenBid

v0.2.20 · live · bid.openli.ai

AI-native Knowledge & Proposal Intelligence. A bid is a governed workspace; the agent works it end-to-end — intake to submission — grounded in the firm’s entitled knowledge, cited, governed and measured at every step. Forked from OpenSOP. Reference scenario: Ashford Partners, a 2,000-consultant management consultancy.

Unified 3-runner (OpenLi Codex / OpenAI Codex SDK / Claude Agent SDK) on one governed contract
Hybrid RAG (dense pgvector + BM25-lite, RRF) over the corpus · a derived expertise graph
4-dimension access governance + conflict walls + freshness · governance-leak rate 0
Per-step evaluation gates (lifecycle 9/9, no-regression) · in-workspace audit trail

Open the platform Product page GitHub

Compliance & assurance alignment

Every product in the OpenLI family inherits the same compliance posture from the OpenLI Codex foundation. One security review, one audit story, one governance model across healthcare, finance, pharma and travel.

EU AI Act UK AI governance guidance UK GDPR EU GDPR Data Protection Act 2018 ISO/IEC 27001 ISO/IEC 42001 SOC 2 control mapping NIST AI RMF NHS DSPT readiness FCA Consumer Duty alignment MiFID II suitability ICH-GCP & 21 CFR Part 11 alignment

Roadmap

The portfolio scales one foundation into a tree of sector products. Branches launch when the foundation is ready and the buyer signal is strong.

2026 H1 — Foundation consolidation

OpenLI Codex v0.7. Healthcare cluster live (HIE, Integrai, IRIS CoPilot, OpenTriage, DMM). OpenCT live for UK Corporation Tax filing. GSJ Platform live with Great Scenic Journeys. OpenTrials live at ct.openli.ai for clinical trial RBQM and central statistical monitoring.

2026 H2 — Pipeline early-access programmes open

OpenInvestor early access opens for banks, private banks and IFA platforms. OpenTrials adds the EDC connector wave and submission packaging. All built on the same OpenLI Codex foundation.

2027 — Cross-sector marketplace

Reusable AI skills, integration packs and governance policies shared across the portfolio. Additional verticals (audit, public sector, energy) on the same foundation.