CNCK0NSULTAI-Truthuni0naiHackatonipIII ↗k0nsult.dev ↗ dla botówDomeny osobne, spięte siecią CNC + kernelem.
K0NSULT // ai-truth/ipIII
k0nsult.cloud / ai-truth / ipIII / enterprise-trust / en

Enterprise Trust & Evidence Governance

One page collecting everything a CISO and procurement team need to evaluate ipIII before a pilot: PoC scope, data handling principles, retention, RBAC, audit log, the CVD program, legal documents (RoE/DPA/NDA), known limitations, deployment modes and the SLA model. Doctrine claim ≤ proof: what works is labeled LIVE, what is planned — ROADMAP.

Why this page exists. Security and procurement teams at banks, insurers and public administration bodies do not have time to browse ten scattered subpages before a first conversation. This pillar collects them in one place, with links to the details, and states plainly what is proven today versus what is still planned. Zero certifications we do not hold. Zero guarantees we cannot keep.
Evidence-first governance: 8 trust pillars, one entry-point document.

ipIII is an evidence and orchestration layer on top of existing scanners and security processes — today at the stage of a demonstrable MVP (import → incident → evidence package → retest → close, with JWT/RBAC/audit on a live PostgreSQL instance). This page walks the CISO and procurement team through eight areas, each with its own dedicated subpage containing details, evidence and downloadable documents.

EVALUATION PATH: 1. PoC scope2. Data & retention3. RBAC & audit4. CVD5. Legal (RoE/DPA/NDA)6. Limitations7. Deployment mode8. SLA

1. Pilot (PoC) scope — what we test and what we prove

Every ipIII pilot begins with a written agreement on scope: which systems (or their synthetic equivalents) are covered, the time horizon, the client-side point of contact, and the success criteria that decide whether the engagement moves to a production phase. The PoC is not an offensive penetration test — it verifies whether the evidence layer (importing scanner output, normalizing incidents, building the evidence package, generating a board pack) works on the client's data within their integration environment.

What is in scope

Import from existing scanners (Burp/ZAP/Nessus/CSV), normalization into incidents, an evidence package with a checksum, a PDF board pack, and the retest→close path.

What is NOT in scope

ipIII does not perform offensive scanning, does not generate or execute payloads, and does not replace a penetration test commissioned to a red-team.

PoC exit criteria

Agreed jointly before the start: number of reports processed, time from import to evidence package, completeness of the audit log. The output is an evidence-backed report, not a declaration of production readiness.

A full description of scope, a sample timeline and a preparation checklist for the client team is available in the Procurement Pack — a document intended for handoff to the purchasing team alongside the proposal.

2. Data, synthetic inputs and retention

During the pilot phase, ipIII works by default on synthetic or anonymized data. If the client wants to test real scanner reports containing personal or business-sensitive data, this requires a separate data processing agreement (DPA) and an explicit legal basis. Rules on data minimization, retention and deletion after the PoC ends are described in detail on a dedicated page.

AreaPoC ruleStatus
Input data typeSynthetic / anonymized scanner reports (by default)LIVE
Personal / sensitive dataOnly after a signed DPA and an established legal basisconditional
Evidence package retentionPer the schedule agreed in the RoE; deletion on request after the pilot endsLIVE
Storage locationPostgreSQL managed by K0NSULT (EU region); details in Data ProcessingLIVE
Export / right to deletionClient can request export of the evidence package and data deletion after the engagement endsprocedure ROADMAP

Full description of legal bases, controller/processor roles and the data flow map → Data Processing.

3. RBAC and audit log

Access to ipIII is role-controlled (RBAC) — each user is assigned a role (e.g. analyst, reviewer, administrator) that limits which actions they can perform: import, incident review, evidence package generation, close with evidence. Every action is recorded in an audit log with a timestamp, user identifier and operation type — on a live PostgreSQL instance, not in text files.

RBAC LIVE

Roles and permissions enforced at the API level (JWT + authorization middleware).

Audit log LIVE

Every operation on an incident and evidence package recorded with actor and timestamp.

OIDC / SSO ROADMAP

Today JWT is issued locally; identity federation (Keycloak/JWKS) is planned — see Known Limitations.

Multi-tenant RLS ROADMAP

Today one organization per instance; Row-Level Security for multiple clients on a single instance is planned.

4. Coordinated Vulnerability Disclosure (CVD)

ipIII runs a responsible vulnerability disclosure program. Security researchers and partners can report a discovered vulnerability via the form on the /zglos page or the channel described in the Security Pack. Rule: zero publication of exploit details before an agreed fix, communication limited strictly to what was tested and what evidence of a successful fix was provided — no payloads or attack instructions in public content.

Details of the program, the reporting channel and the stated response time → Security Pack.

5. Legal documents: RoE, DPA, NDA

Every engagement that includes a testing component (PoC, pilot, workshop) is preceded by a complete set of documents:

These are decision-support tools for the contracting parties — they do not replace review by the client's own legal counsel or K0NSULT's. Full set of templates and a legal checklist → Procurement Pack.

6. Known limitations — stated plainly

ipIII does not pretend to be a bank-grade system. A register of seven known limitations (identity federation, mTLS, formal PAdES/TSA signature, multi-tenant, live CVE freshness, SIEM connector, Legal Trigger Engine boundary) is documented in full, with named risk, mitigation and closure priority (P0/P1) on a separate page. None of these limitations is hidden or presented as already resolved.

Why this matters for procurement. Evaluating a security vendor that openly lists its own gaps with a remediation priority is easier than evaluating a vendor that stays silent. Full register → Known Limitations.

7. Deployment modes

ipIII can run in different topologies depending on the client's data residency and integration requirements — from a shared instance (multi-tenant, ROADMAP) to a dedicated instance per client (today the default model for banking/institutional pilots). The choice of mode affects SLA, cost and deployment timeline.

ModeDescriptionStatus
Dedicated instance (single-tenant)Separate database and environment per client, physical data isolationLIVE
Shared instance (multi-tenant + RLS)One instance, logical separation via Row-Level SecurityROADMAP
On-prem / client VPCDeployment on client infrastructure instead of K0NSULT hostingROADMAP

Full comparison of topologies, network and cost requirements → Deployment Modes.

8. SLA and support

The support model describes contact channels, stated response times by severity, and support hours available during and after the pilot phase. Detailed response time matrix, escalation and maintenance windows → SLA / Support.

Vendor Risk assessment

For GRC and risk management teams that must classify ipIII as a third-party vendor (third-party risk assessment), a set of answers to standard vendor questionnaire questions is prepared: data location, subprocessors, business continuity, incident response plan. This is an input to the client's own risk assessment process, not a substitute for it → Vendor Risk.

The eight pillars at a glance

8
trust areas
PoC, data, RBAC, CVD, legal, limitations, deployment, SLA
4
pillars LIVE today
PoC evidence flow, RBAC, audit log, RoE/NDA/DPA as a process
4
pillars with ROADMAP elements
multi-tenant RLS, OIDC/SSO, on-prem, data export/deletion
7/7
limitations disclosed openly
zero hidden gaps — see known-limitations

What this page does NOT mean

This is not a certificate or a compliance declaration. Enterprise Trust & Evidence Governance is a map of documents and evidence that a CISO/procurement team consults before deciding on a pilot. Assessing whether ipIII meets the requirements of a specific institution (bank, insurer, public authority) remains the responsibility of the client's own vendor risk management process — the materials here are decision-support, not a replacement for it.
"100%" here means evidence coverage, not impenetrability. Every element marked LIVE has code, a test and an endpoint that can be checked. An element without such evidence is marked ROADMAP — never presented as a finished feature.
Ethical and legal boundary. ipIII is a defense and compliance tool (GRC/blue team), not an offensive one. All testing takes place exclusively on synthetic data or within the boundaries of a written Rules of Engagement. The Legal Trigger Engine and legal materials are decision support, not legal advice — qualification of any document before submission to a regulator is the responsibility of the client's own legal counsel or law firm.

Next step

View Trust Center

Full overview of trust policies, certifications in progress and LIVE evidence.

→ /ai-truth/ipIII/trust-center

Download the Procurement Pack

PoC scope, legal checklists, RoE/DPA/NDA templates ready for the purchasing team.

→ /ai-truth/ipIII/procurement

Known Limitations

Register of seven limitations with risk, mitigation and priority — zero hidden gaps.

→ /ai-truth/ipIII/known-limitations

Related: data processing and legal bases → /data-processing · security program and CVD → /security-pack · deployment modes → /deployment-modes · SLA and support → /sla-support · vendor risk assessment → /vendor-risk.