Ontology-Driven Agent Control: The Semantic Foundation of Harness Engineering

From Agent Hype to "Uncontrollable"

In 2024-2025, agents became the primary form of enterprise AI deployment, capable of autonomous planning, tool use, and multi-step execution. However, in real business scenarios they exhibit consistent failures: inaccurate terminology, divergent reasoning logic, and outputs that violate enterprise rules — "confidently doing the wrong thing" at critical nodes. The root cause is not model weakness but the lack of a "rule-aware structure": agents know how to act but not the industry boundaries or enterprise decision rules.

This context gave rise to Harness Engineering in Q1 2026, aiming to build a complete constraint, feedback, and control system so agents can act autonomously without crossing business boundaries. The industry converged on three technical pillars: architectural constraints to limit behavioral boundaries, context engineering to manage information input, and feedback loops to verify output quality. Yet most implementations remain at the engineering-assembly level — writing prompts, listing rules, building workflows, adding permission controls. These work for simple scenarios but fail in complex business because constraints are "externally added," semantics are "implicit," and compliance relies on the model's unreliable understanding of rules.

This article proposes a different path: using ontology as the semantic infrastructure for the Harness system, making constraints derive from business structure itself rather than engineering configuration. Ontology explicitly models business semantics, providing a unified structured foundation for the three pillars.

Redefining the Problem: "Safe and Controllable" Is a Multi-Dimensional Engineering Proposition

Before detailing the solution, the article clarifies the dimensions of "safe and controllable execution" via a table:

Permission & Isolation : Who can do what? Can data cross boundaries? → RBAC/ABAC, API gateways, data sandboxes.

Behavioral Constraints : Where are the agent's reasoning and invocation boundaries? → Prompt constraints, tool whitelists, ontology modeling.

Audit & Traceability : What was done? Can the decision process be reconstructed? → Operation logs, decision-chain tracing, explainability frameworks.

Exception Handling : How to degrade and roll back on errors? → Circuit breakers, human review nodes, idempotent design.

Result Verification : Does output comply with business rules? → Rule engines, formal verification, ontology constraint validation.

Compliance Alignment : Does it meet industry regulatory requirements? → Compliance knowledge bases, approval-flow integration, auditable reports.

The ontology-driven approach primarily addresses Behavioral Constraints and Result Verification , with substantive contributions to Audit & Traceability . It serves as the semantic infrastructure layer within a complete safety system, not a replacement for other engineering measures.

Architectural Constraints: From "External Fence" to "Internal Skeleton"

Engineering constraints face three structural difficulties in complex business: rule count grows with complexity, inflating maintenance costs; rules expressed in natural language suffer from model interpretation uncertainty and bypass risk; semantic relationships between rules and business objects are implicit, preventing cross-scenario reuse.

Ontology differs by defining the agent's behavioral space upfront. Constraints become an internal skeleton built into business structure, not an external fence. Business rules are explicitly modeled as queryable, verifiable structures rather than living in prompts. Tools are managed uniformly at the ontology layer — what an agent can use, how to trigger, and execution flows are dictated by structure, not model discretion.

Constraint enforcement occurs after agent output but before operation execution: the system retrieves the agent's execution intent and compares it against the ontology. If predefined rules are violated, the action is rejected and the agent must re-reason. Validation relies on traceable nodes and relationship paths in the structure, yielding deterministic conclusions independent of model understanding.

Context Engineering: From "Supplementing Memory" to "Restructuring Memory"

Agents frequently "lose memory" in long tasks, repeatedly asking for basic information or losing context at critical nodes. The root cause is linear text stacking without structure. External memory, context compression, and window expansion only alleviate the symptom.

Ontology takes a different entry point: enterprise data, processes, and relationships are inherently a complex association network. Ontology makes these relationships explicit, building a queryable, evolvable business semantic network. This structure brings three concrete improvements:

Precise Retrieval Replaces Full Injection : Before agent startup, the cognitive engine extracts a task-relevant semantic subgraph from the ontology, dynamically injecting only the most relevant context into the reasoning context, eliminating context overflow and irrelevant noise at the source.

Consistency Assurance : Business knowledge is managed by a unified semantic network; expired information, conflicts, and redundancies are systematically handled. Agents always reason on the latest consistent information, not on potentially outdated prompt fragments.

Cross-Task Reuse : The same semantic structure serves different agents across tasks, eliminating per-scenario context reorganization. The ontology is a continuously maintained "business map" on which agents act, rather than starting from zero each task.

Additionally, ontology-driven context engineering resolves the long-standing split between symbolic reasoning and LLM reasoning. Traditional knowledge graphs are explainable but limited in coverage; pure LLMs are flexible but uncontrollable. Ontology provides a collaboration framework: where structure covers, ontology gives deterministic constraints; where structure doesn't cover, LLMs can fill in but conclusions are explicitly tagged with confidence status. Both play to their strengths, preserving accuracy and flexibility.

Feedback Loops: From "Subjective Evaluation" to "Traceable Verification"

Mainstream feedback mechanisms introduce an "evaluator" model to judge another model's output. This has limited effect because models are easily persuaded by superficially reasonable results and lack business-level correctness judgment.

Ontology offers a more direct path: objective verification based on structure, not subjective evaluation. Many enterprise judgments are essentially rule-based — whether a quota is exceeded, preconditions satisfied, process logic followed. These don't require "understanding"; they can be directly verified. Once encoded in ontology, every agent output can be automatically compared against business constraints, yielding a binary pass/fail.

Not all judgments can be formalized. "Hard constraints" suit ontology verification; "soft constraints" (business conventions, anomalous motives, definition of "major change") do not. For soft constraints, ontology must combine with LLM evaluation or human review nodes — complementary, not mutually exclusive.

Another value of the feedback loop is continuous ontology evolution. The ontology-agent relationship is not one-way constraint: ontology defines agent capability boundaries; agents encounter massive business data during real execution, which feeds back into ontology — identifying uncovered concepts and relationships, marking frequent error locations in reasoning paths, guiding ontology correction and iteration. This continuous evolution loop distinguishes it from traditional "build-and-forget" knowledge graphs.

Explainability is another critical value. Every verification conclusion has explicit structural basis; every judgment traces to specific rule nodes and relationship paths. Regulatory audits need "system can prove compliance," not "model thinks it's compliant." This gap is decisive in high-compliance industries like finance, industrial, and railway.

From Technical Control to Business Control: Knora's Implementation Path

1. Constraints Come from Business Itself, Not Engineering Assembly

Traditional Harness solutions make agents "more stable," but stability boundaries are engineering boundaries — rules written by humans, constraints configured, compliance maintained via prompts. The ceiling is clear: once business complexity exceeds engineering maintenance capacity, constraints fail.

Ontology-driven approach achieves a different outcome: constraints derive directly from business structure. Enterprise objects, relationships, and rules are uniformly modeled as ontology. When executing tasks, agents don't just receive context and generate; they must first enter this semantic layer, completing reasoning and decision-making within predefined business relationships. They act within a defined "business map," not in free play.

This shift brings three substantive transformations: agents become business-semantic participants, not mere execution tools; systems reuse constraints on a unified semantic foundation instead of rebuilding per scenario; business rule layers remain stable during model iteration, avoiding drift with model versions. This is the leap from "technical control" to "business control."

2. Knora's System Architecture

YueDian Technology's Knora platform implements the methodology as a layered collaborative architecture.

Ontology Layer (Knowledge Foundation)

Bottom layer stores ontology models in a Labeled Property Graph (LPG) . At the meta-schema level, five core concepts are defined:

Entity : Business objects (e.g., work orders, equipment, personnel).

Relation : Semantic connections between entities.

Event : Meaningful state changes in business.

Action : System-executable operations, including trigger conditions and parameter constraints.

Logic : Execution flows defined by a DAG orchestration engine.

Key design: Traditional ontologies address "what is" classification; Action and Logic transform ontology from static knowledge description into executable business specifications that directly drive agent tool calls and process orchestration. Tools are defined at the ontology layer, not the agent layer — what tools an agent can use, how to trigger, and execution flows are all decided by ontology. This layer is the system's knowledge foundation, continuously evolvable.

Cognitive Engine Layer (Translation & Arbitration)

Bridges ontology and agent. Two core roles:

Before agent startup: reads domain knowledge relevant to the current scenario from ontology — entity relationships, business rules, available tools — and injects into the agent's reasoning context, so the agent knows both "how this domain operates" and "which tools I can use and how."

After agent generates results: retrieves conclusions and validates against ontology constraints. If rules are violated, rejects and forces re-reasoning.

Agent Execution Layer (Task Executor)

Autonomous intelligent agent product layer. Receives user tasks, calls tools, generates results, completes execution. Pure executor — tool origin, usage, boundaries are not self-determined but defined by ontology layer and passed via cognitive engine.

Data flow: User task → Cognitive engine extracts relevant knowledge from ontology and injects into context → Agent reasons and executes in this context → Results return to cognitive engine for ontology validation → Only after passing is final output produced. Ontology does not interact directly with agent; it acts through the cognitive engine intermediary.

Manufacturing Example: Work Order Quantity Change Requires Approval

Complete execution chain:

User intent: "Change work order WO-2026-0312 planned quantity from 500 to 800."

Cognitive engine queries ontology (LPG graph traversal):

Node: WO-2026-0312, type WorkOrder, status "Issued/Not Started".

Attribute: Change magnitude 60%, exceeding ontology-defined 20% approval threshold.

Action rule: Over-threshold change triggers condition requiresApproval = true.

Relationship chain:

WorkOrderChange → approvedBy → [BOMEngineer, ProductionManager]

.

Constraint validation: Current operation lacks approvedBy relationship → validation fails.

System response:

Blocks direct write.

Generates structured error report with specific violated rule nodes and relationship paths.

Auto-creates approval task, routes to corresponding approvers.

Writes operation intent to audit log with status "Pending Approval".

After approval: Adds approvedBy relationship, re-validates → passes → executes write.

3. Automated Modeling: Layered Processing & Human-Machine Collaboration

Enterprise ontology cold-start cost is a real challenge. Manual construction of high-coverage ontology traditionally takes weeks or months with professional knowledge engineers. Knora's strategy is not "fully automatic in one step" but layered processing, confidence-driven, human-machine collaboration fallback. Structured, pattern-strong tasks (field-to-ontology-attribute semantic mapping, auto-generating data governance DAG flows from ontology metadata) are automated; tasks involving business semantic judgment and concept boundary definition are not forced. System assigns confidence scores to each auto-generated result: high confidence executes directly, medium confidence recommends human confirmation, low confidence enters review queue. Human intervention targets only truly ambiguous parts. Each human confirmation or correction feeds back to continuously optimize subsequent auto-modeling capability. As industry templates accumulate and domain samples adapt, human intervention ratio gradually decreases.

4. Real-World Deployment Results

Knora has production deployments in energy transport, electronics manufacturing, finance, and security.

Energy Transport : Railway comprehensive inspection report generation agent automatically constructs indicator systems and generates reports. Work that originally required 30 people / 7 days to organize data and write reports becomes 3 people spending one day verifying data, agent completing in 30 minutes — efficiency improvement over 70x . Deployed as "digital employees" across dozens of scenarios: safety hazard identification, maintenance plan generation, emergency plan generation, emergency resource scheduling.

Electronics Manufacturing : Core scenarios like quality traceability and defect analysis transform experience-dependent manual processes into precise knowledge-driven automated digital business flows.

Conclusion

AI's entry into enterprise is undergoing a fundamental fork. One path continues stacking tools, accumulating prompts, doing integrations — agents run but no one truly knows where they'll go or what results they'll produce. The other path builds business structure first, letting agents act on a clearly defined semantic map — they know boundaries, rules, and the basis for every decision.

Short-term cost difference is small, but after one year, three years, the gap appears at every moment requiring regulatory explanation of decisions, every multi-system collaboration with semantic confusion, every critical node where an agent "confidently does the wrong thing."

Enterprise AI competition is ultimately not between models, but who structures their business knowledge first . Enterprises that close this loop first accumulate not just a system, but a continuously self-evolving business intelligence foundation — becoming more accurate with every agent execution, more complete with every business iteration.

The real moat is never bought tools, but precipitated cognition. Agents will be replaced, models will be iterated, but business cognition沉淀 in ontology will not — that is the reason to start now.