What is the smallest Ordivon Harness that turns a bare model API into a verifiable Agent Run?
Build one thin first-party model/Tool Loop without copying Host Task truth, Runtime process truth, or mature Provider Harness lifecycles.
01 / Current position
A hypothesis is not the current judgment.
The dossier preserves the live research position. Dated articles preserve the complete evidence and argument that changed it.
A Host-local v0 with one Model Adapter, sequential Tool Loop, Assignment-scoped catalog, Runtime bridge, bounded Run state, interruption, and existing Run receipts can solve one real multi-turn workload while remaining materially smaller than mature Provider drivers.
Ready for construction at M2. The complete model-to-work stack, H1–H5 ownership boundary, first falsifier, comparison baselines, acceptance, non-goals, and deletion condition are frozen. No implementation evidence exists yet.
02 / Decision boundary
What keeps this Question alive?
Without a first-party Loop, Ordivon can only use complete external Agent products or one-shot model judgment. Bare APIs, local models, controlled Loop research, and strategic fallback capability remain inaccessible.
Compare one frozen repository workload under one-shot ModelGateway, Ordivon Harness with a bare model API, and one mature Provider Harness; retain the prototype only if it solves a real failure or enables a real model path at acceptable permanent cost.
One-shot or mature Provider Harness paths match correctness, recovery, portability, and cost while no bare-model use case requires the first-party Loop.
03 / Supporting publications
Complete arguments connected to this Question.
3 dated publications currently document this research line.
From Tokens to Work: The Complete Agent Execution Stack
A first-principles walkthrough from token generation to tool execution, verification, TaskOutcome, and graph advancement, compared with OpenAI, Anthropic, Microsoft, Google, and Ordivon's tested boundaries.
31 July 2026 ↗Research reportWhat Survived When Codex and Hermes Replaced Each Other Mid-Task
How one task remained coherent through Codex→Hermes and Hermes→Codex replacement, three injected faults, and materially different provider lifecycles.
31 July 2026 ↗Architecture decisionWhy Ordivon Needs a Harness—but Not a Universal Harness
Why H5 correctly rejected a common Codex/Hermes lifecycle while a separate Ordivon Harness remains strategically necessary for bare model APIs, local inference, and controlled agent-loop research.
31 July 2026 ↗04 / Source discipline
The dossier is an index, not the evidence authority.
Owns the current judgment, next test, and deletion condition.
Own complete dated arguments, limitations, comparisons, and source links.
Own exact code, tests, releases, receipts, and machine evidence.