Can Agent capital judgment remain separate from financial effect authority?
Pressure quantitative research, proposal/decision semantics, signer and credential authority, one-shot dispatch, and venue reconciliation without turning a backtest or model conclusion into permission to trade.
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.
Agent judgment can remain flexible while deterministic Finance/Executor machinery owns exact effect identity, admission, dispatch evidence, and reconciliation under explicit deployment assurance.
Testing at a later boundary than the original QB experiments. Finance now preserves real FinancialResources independently of current risky positions, aggregates component portfolio state with explicit FX/temporal evidence, binds owner-managed capital through replayable CapitalState/Exposure context, and blocks both stale state-changing Decisions and unsupported owner-contribution shortcuts. Context v21 routes only work pulled by current Finance consumers into authoritative obligations while external financial submission remains independently fail-closed.
02 / Decision boundary
What keeps this Question alive?
Real-capital systems fail dangerously when predictive evidence, semantic judgment, API permission, dispatch identity, and final venue consequence are treated as the same state.
Use the current owner-capital and consumer-bound context to evaluate real opportunities without relabeling missing return/history evidence as zero risk or positive contribution. Keep external submission separate until its selected deployment profile, current credentials/signer, one-shot dispatch, and reconciliation prerequisites are explicitly admitted.
A simpler owner-native venue integration preserves the same effect identity, one-shot dispatch, evidence, and reconciliation with less permanent Finance machinery, or the Agent/effect boundary fails under a bounded canary.
03 / Supporting publications
Complete arguments connected to this Question.
5 dated publications currently document this research line.
The Decision Was Valid. The Portfolio Changed.
Why Decision validity and commitment currentness are different facts, using Ordivon Finance's exact portfolio-baseline fence for state-changing proposals.
22 August 2026 ↗Research reportCorrectness Is Not Isolation
How Finance C2-L separated effect correctness from credential and administrator isolation, falsified hostile-root isolation on one machine, and retained a disabled single-machine effect-correctness profile without authorizing live trading.
12 August 2026 ↗Research reportThe Agent Asked for More Evidence. The Answer Became No.
Finance QB6 as a concrete example of evidence-driven non-action: an Agent requests missing read-only evidence, updates its judgment, and terminates before simulation, Decision, Proposal, or financial effect.
12 August 2026 ↗Architecture guideThe Command Succeeded. Did Anything Happen?
A concrete guide to the boundary between local execution, external consequence, and semantic completion, using Runtime's effect kernel and Security's interrupted-consequence experiment.
12 August 2026 ↗Research reportThe Constitution Changed the Agent. That Wasn't Enough.
Why Finance QB3b/QB3c refused to promote an apparently useful Agent constitution after an unchanged sealed replication reversed the result, and what that says about context treatments, priors, and default prompts.
12 August 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.