GitHub - base76-research-lab/TrustPlane: Base76s enterprise

6 min read Original article ↗

TrustPlane

The AI Trust Control Plane for production LLM systems.

🔖 Acquisition opportunity: TrustPlane is available for acquisition. Listing coming soon on Acquire.com. Contact bjorn@base76.se for early inquiries.

TrustPlane gives your legal and compliance team what they need to sign off on AI deployment: a documented risk policy, an immutable audit trail, and automatic escalation to human review when confidence drops.

Powered by CognOS — an auditable epistemic risk engine with published theoretical foundations.

The risk your organization is carrying

Every unverified LLM response in a production system is an undocumented decision. When something goes wrong — a hallucinated figure in a financial report, an incorrect medical reference, a legal citation that doesn't exist — there is no record of what the model was asked, what it answered, or what your system decided to do with it.

That is a liability exposure. In regulated industries, it is a compliance failure.

TrustPlane intercepts every LLM request, scores it for epistemic uncertainty, enforces your risk policy, and logs the decision — before the response reaches your users.

How it works

Every request is scored using the epistemic trust formula:

  • p — prior confidence for this request type
  • Ue — epistemic uncertainty (what the model doesn't know)
  • Ua — aleatoric uncertainty (irreducible noise in the domain)

The score determines the outcome:

Trust score Decision Action
Above threshold PASS Request forwarded
Threshold − 0.2 REFINE Warning headers, logged
Threshold − 0.4 ESCALATE Webhook triggered, human notified
Below minimum BLOCK Request rejected, trace saved

Every response carries the decision in its headers:

X-TrustPlane-Score: 0.8731
X-TrustPlane-Decision: PASS
X-TrustPlane-Trace-Id: tr_a3f91c2d44b1
X-TrustPlane-Policy: enterprise_v1

Evidence base

The scoring model is open source, auditable, and grounded in published theory.

OSS core: cognos-proof-engine — the scoring engine is MIT-licensed and independently verifiable. The formula, thresholds, and policy engine are all inspectable.

Published research: The Field-Node-Cockpit (FNC) framework underlying the model has been through peer review. Full publication record at Applied AI Philosophy.

Falsifiability: A system calibrated to target_risk: 0.3 escalates measurably more than one set to 0.7. This is testable against your own request logs — not a marketing claim.

If your security team, legal counsel, or compliance officer needs to understand how the scoring works, there is source code and published theory to show them.

Capabilities

Pluggable LLM backend

One control plane. Every provider. Switch without changing application code.

provider: ollama          # On-premise, air-gapped
provider: openai          # GPT-4o, o1, o3
provider: anthropic       # Claude Sonnet, Claude Opus
provider: groq            # High-throughput inference
provider: cerebras        # Low-latency inference

Fallback providers configurable for zero-downtime failover.

Multi-tenant isolation

PostgreSQL schema-per-tenant. No data shared between organizations.

tenant_acme.traces
tenant_globocorp.traces
tenant_healthsystem.traces

Human oversight triggers

Webhook dispatch on ESCALATE and BLOCK. Your incident management system receives the signal before your users receive the response.

{
  "trace_id": "tr_a3f91c2d44b1",
  "decision": "ESCALATE",
  "trust_score": 0.31,
  "tenant_id": "acme",
  "timestamp": "2026-03-02T14:22:11Z"
}

Audit trail

Complete trace history for every request. Exportable as CSV or EU AI Act Article 13 PDF report.

curl -H "X-API-Key: your-key" \
  "https://your-gateway/v1/audit/export?format=pdf&from=2026-01-01" \
  -o compliance_report.pdf

In air-gapped deployments, audit trail integrity is enforced through physical and network isolation — the same model used in defense, healthcare, and public sector infrastructure. No external process can reach the database. For deployments requiring additional guarantees, WAL archiving to write-once storage can be added above TrustPlane.

Deployment options

Self-hosted via Docker Compose (air-gap compatible) or SaaS. Your data never has to leave your environment.

Claude Code & Anthropic API

TrustPlane integrates with Anthropic in two directions.

Claude as LLM backend — any Claude model behind the trust-scoring gateway:

provider: anthropic
model: claude-sonnet-4-6
api_key: ${ANTHROPIC_API_KEY}
target_risk: 0.3

TrustPlane as MCP server for Claude Code — expose trust verification as tools Claude Code can call during its own reasoning:

{
  "mcpServers": {
    "trustplane": {
      "command": "python",
      "args": ["/path/to/Cognos-enterprise/mcp/server.py"],
      "env": {
        "COGNOS_BASE_URL": "http://127.0.0.1:8788",
        "COGNOS_API_KEY": "your-key"
      }
    }
  }
}

Full setup guide: mcp/CLAUDE_CODE_SETUP.md

Quickstart

git clone https://github.com/base76-research-lab/Cognos-enterprise.git
cd Cognos-enterprise
cp .env.example .env        # Set COGNOS_PROVIDER + API key
docker-compose up

curl -X POST http://localhost:8788/v1/chat/completions \
  -H "X-API-Key: test-key" \
  -H "X-Cognos-Tenant: demo" \
  -H "Content-Type: application/json" \
  -d '{"model":"ollama/llama3.2:1b","messages":[{"role":"user","content":"Hello"}]}'

Architecture

┌─────────────────────────────────────────────────────┐
│                  Your Application                    │
└─────────────────────────┬───────────────────────────┘
                          │
                          ▼
┌─────────────────────────────────────────────────────┐
│                    TrustPlane                        │
│                                                     │
│   Auth & RBAC  →  Trust Scoring  →  Policy Engine   │
│                   C=p(1−Ue−Ua)     PASS/REFINE/     │
│                                    ESCALATE/BLOCK    │
│                                         │           │
│   Rate Limiting    Tenant Isolation     │           │
│   (Redis)          (per-schema PG)      │           │
│                                         ▼           │
│              ┌──────────────────────────────────┐   │
│              │         Provider Router           │   │
│              └───┬──────┬──────┬──────┬─────────┘   │
└──────────────────┼──────┼──────┼──────┼─────────────┘
                   │      │      │      │
                Ollama  OpenAI  Anthropic  Groq/Cerebras

Full diagram: docs/ARCHITECTURE.md

EU AI Act compliance

Designed for organizations operating high-risk AI systems under EU AI Act Annex III.

Article Requirement How TrustPlane addresses it
Art. 9 Risk management system Continuous epistemic scoring on every inference
Art. 12 Record-keeping Immutable trace log with trace IDs and timestamps
Art. 13 Transparency Automated PDF reports with full attestation summary
Art. 14 Human oversight Webhook escalation before downstream consequences

Full compliance guide: docs/EU_AI_ACT.md

→ Technical whitepaper — theory, implementation, limitations, EU AI Act mapping

Licensing

TrustPlane is available for licensing to organisations and consultancies that need to deploy EU AI Act compliance infrastructure without building from scratch.

Three models available:

  • Single organisation — deploy internally, own the infrastructure
  • Reseller / consultancy — white-label for client deployments
  • Research / pilot partnership — co-develop for a specific domain (healthcare, finance, public sector)

→ licensing/LICENSE_OFFER.md — full terms and pricing

Contact: bjorn@base76.se

Use case scenarios

Detailed deployment scenarios with architecture diagrams:

Pricing

Free

Open-source core. 1 tenant. 100 requests/day. CSV export. → cognos-proof-engine (MIT)

SaaS — from €999/month per tenant

Hosted. Zero infrastructure. Full enterprise feature set.

Self-hosted license — from €25,000/year

Your infrastructure. Air-gap compatible. Full source access.

Need a human-reviewed compliance report? Base76 Research Lab offers provider-facing EU AI Act Deep Audits - Article 11, 13, 14, 15 mapping, Annex III triage. From €20K. https://base76research.com/deep-audit

Enterprise consulting

Architecture review, policy calibration, compliance documentation, ongoing support. For healthcare, legal, finance, and public sector deployments.

Contact: bjorn@base76.se

Free vs Enterprise

Feature Free Enterprise
Trust scoring (PASS/REFINE/ESCALATE/BLOCK) ✓ ✓
All LLM providers ✓ ✓
Trace history + CSV export ✓ ✓
Webhooks 1 endpoint / 100 events/day Unlimited
Tenants 1 Unlimited
Rate limit 100 req/day Configurable
PDF audit reports (EU AI Act) ✗ ✓
Fallback providers ✗ ✓
Custom RBAC roles ✗ ✓
Session memory ✗ ✓
Token compression ✗ ✓
SLA + support ✗ ✓

Built on open source

TrustPlane is the production layer on top of cognos-proof-engine (MIT). The scoring engine is open and auditable. TrustPlane adds multi-tenancy, auth, webhooks, audit exports, and commercial support.

Related:

Base76 Research Lab — Sjöbo, Sweden