The AEPT-AI knowledge base (v5.0.0 Prometheus, 34 ontologies, 34 agents, 92 skills, 65 policies, 36 scripts, 10 team templates, 25 commands) represents a substantial and monetizable intellectual property foundation. Governed by the Five-Layer Stack (Capability, Projection, Protocol/Transport, Runtime/Governance, and Persistence planes), this analysis maps each ontology and policy to nine commercially viable product candidates, each targeting a distinct segment of the enterprise AI, security, and compliance software market.
The highest-priority opportunity is AgentShield (P1): a runtime AI agent security platform implementing OWASP Agentic AI Top 10 (ASI01-ASI10), CSA MAESTRO L1-L7, and NIST AI RMF as enforceable controls. No competitor currently offers this coverage. The regulatory tailwind (EU AI Act Aug 2026; prompt injection +540%; 79% LLM vulnerabilities unresolved) creates an urgent, funded buyer at the CISO level. The platform is now governed by 34 deployed agents and a 3-tier write lease (G-067), hardening the defensible moat considerably since v2.1.0.
The fastest path to revenue runs through ProspectIQ (P5, readiness 93%) and ContextVault (P6, readiness 87%). Both products are executable from existing AEPT-AI components within 2-4 months and address well-understood pain points in enterprise AI design and memory governance. A live UCG client engagement validates ProspectIQ's production readiness.
The three highest-reuse components (ONT-CALIBRATION-010, ONT-GRAPH-RAG-007, fact-checker) form a shared intelligence substrate present across four or more products. Building this substrate first maximizes development leverage across the portfolio and creates compounding switching costs for enterprise customers.
The AEPT-AI v5.0.0 Prometheus platform inventory comprises 34 ontologies spanning AI security, regulatory compliance, inference architecture, neuro-symbolic reasoning, and evidence provenance. The 34 specialized agents cover the full engineering lifecycle from planning through deployment and monitoring, organized across five functional teams (core engineering, audit, RPA, and BHT). Seven skills encapsulate compound workflows including GBP auditing, Folio workspace management, and the Skill Schema Registry. The 65 governance policies encode institutional knowledge as enforceable machine-readable controls, governed by the Five-Layer Stack (G-064). Together these assets form the foundation for all nine product candidates described in this analysis.
Click any column header to sort. Products ranked by composite score: readiness (40%), CAGR (30%), market size (20%), time-to-revenue (10%).
| Rank | ID | Product | Category | Market ($B) | CAGR | Readiness | Time to Revenue | Competition |
|---|---|---|---|---|---|---|---|---|
| #1 | P1 | AgentShield | Security | $13.5B | 22.3% | 75% |
6-12 mo | None |
| #2 | P2 | ComplianceGraph | Compliance | $22.6B | 17.1% | 68% |
6-9 mo | None |
| #3 | P4 | RAGScore | AI Quality | $7.5B | 38.0% | 83% |
3-6 mo | Partial |
| #4 | P3 | ThreatPulse | AI Infrastructure | $29.6B | 20.15% | 82% |
4-8 mo | Fragmented |
| #5 | P6 | ContextVault | Security / Memory | $8.3B | 20.0% | 87% |
3-6 mo | No Direct |
| #6 | P5 | ProspectIQ | Analytics / GTM | $11.7B | 7.7% | 93% |
2-4 mo | Adjacent |
| #7 | P9 | DevForge | Developer Tooling | $11.1B | 23.65% | 73% |
6-12 mo | Point Solution |
| #8 | P8 | IngestPro | Data Infrastructure | $8.5B | 14.2% | 83% |
3-6 mo | Adjacent |
| #9 | P7 | NeuroForge | AI/ML Infrastructure | $2.6B | 33.32% | 58% |
12-18 mo | None |
| #10 | P10 | AegisMCP | Security / Infrastructure | $5.2B | 31.0% | 85% |
2-4 mo | No Local Equal |
Components appearing in 3+ products form the shared platform substrate and command priority build investment.
| Component | P1 Shield |
P2 Comply |
P3 Threat |
P4 RAG |
P5 Prospect |
P6 Vault |
P7 Neuro |
P8 Ingest |
P9 Dev |
Total |
|---|---|---|---|---|---|---|---|---|---|---|
| ONT-CALIBRATION-010 | - | ✓ | - | ✓ | ✓ | ✓ | - | - | - | 5 |
| ONT-GRAPH-RAG-007 | - | ✓ | - | ✓ | ✓ | - | - | ✓ | - | 4 |
| fact-checker | - | - | - | ✓ | ✓ | - | - | - | - | 4 |
| ONT-INFERENCE-ARCH-014 | ✓ | - | ✓ | - | - | ✓ | ✓ | - | - | 4 |
| security-auditor | ✓ | - | ✓ | - | - | ✓ | - | - | - | 3 |
| G-037 (Circuit Breaker) | ✓ | - | ✓ | - | - | ✓ | - | - | ✓ | 3 |
| ONT-NEUROSYM-009 | - | - | - | - | - | - | ✓ | - | - | 3 |
| researcher | - | - | - | ✓ | ✓ | - | ✓ | ✓ | - | 3 |
| ONT-SUPPLY-CHAIN-001 | - | - | ✓ | - | - | - | ✓ | ✓ | - | 3 |
| G-025 (Court Evidence) | - | ✓ | - | - | ✓ | ✓ | ✓ | ✓ | ✓ | 3 |
| ONT-MCP-SECURITY-015 | ✓ | - | ✓ | - | - | - | - | - | - | 2 |
| memory-keeper | - | - | - | - | - | ✓ | - | - | - | 1 |
Darker shade = higher reuse count (0 to 5). White text on levels 3+.
Assumptions: avg deal close 9 months; sales ramp 6 months; expansion revenue 22% annually; churn 8% annually; bundle revenue uplift 35%.
Build the shared platform substrate: ONT-CALIBRATION-010 confidence layer, ONT-GRAPH-RAG-007 knowledge engine, and G-037 circuit breaker. Stand up ProspectIQ (P5) beta with the existing /prospect command tree and researcher + fact-checker pipeline. Target: 3 pilot customers signed.
Launch ProspectIQ free tier with 20 flash report limit. Package ContextVault (P6) memory governance offering using the existing memory-keeper agent + MEMORY.md architecture. Begin RAGScore (P4) developer beta on HuggingFace with CONFIDENT mode benchmark. Target: $120K first ARR contracted.
Complete AgentShield (P1) OWASP ASI compliance self-assessment. Submit for third-party certification. Begin ComplianceGraph (P2) EU AI Act readiness pilot with 2 Fortune 500 compliance teams. Target: 10 customers in pipeline, $480K ARR signed or committed.
The AEPT-AI knowledge base is not merely documentation: it is a product engine. The 34 ontologies, 32 agents, and 65 governance policies represent compounded research across AI security, regulatory compliance, inference architecture, neuro-symbolic reasoning, and evidence provenance. The v5.0.0 Prometheus Five-Layer Stack (Capability, Projection, Protocol/Transport, Runtime/Governance, and Persistence planes) elevates every product candidate's architectural defensibility beyond what any competitor can replicate. Each of the nine product candidates derives its moat from components assembled in depth and breadth that no competitor has matched.
The optimal sequencing is: build the shared substrate first (ONT-CALIBRATION-010, ONT-GRAPH-RAG-007, G-037), ship ProspectIQ (P5) in 2-4 months for revenue validation (93% ready, UCG engagement live), then deploy AgentShield (P1) with OWASP ASI certification as the flagship enterprise offering. This sequence minimizes time to first revenue while building toward the highest-value, most defensible position in the portfolio.
Total addressable market across all ten products: $120.6B. Target capture at 2% penetration: $2.41B revenue potential. The 2026 regulatory supercycle (EU AI Act Aug 2026, BOD 25-01, DORA, NIS2) creates an urgency window that strongly favors early movers with verifiable, standards-aligned implementations. The CAGR weighting across the portfolio averages 22.0%, indicating durable long-term growth in each segment.
Prompt injection attacks increased 540% year-over-year. 79% of LLM vulnerabilities remain unresolved. The OWASP Agentic AI Top 10 was released December 2025 with no existing product implementing all 10 controls as enforceable runtime policies. The window for a first-mover certification advantage in the AI agent security market is 6-12 months before incumbents respond with partial coverage. ProspectIQ (93% ready, UCG live) and ContextVault (87% ready) can generate early revenue while AgentShield completes its certification process. The BHT platform (7 sub-agents, 8 active targets) accelerates ThreatPulse's readiness to 82% — now firmly in the green tier — compressing its Y3 launch to Y2.