A router should produce evidence, not just an agent name

In many agentic systems, a request arrives, candidate agents or tools receive scores, and the highest-scoring option wins. That is enough to execute a workflow, but it is not enough to explain or govern one.

When delegation fails, an operator needs to reconstruct the evidence considered, verify that the selected candidate was permitted and available, understand why alternatives were rejected, and determine whether the system should have clarified, abstained, or escalated instead.

Evidence is not eligibility. A score is not a probability. Probability is not utility. Policy is not intent. Explanation is not provenance.

A decision-provenance lifecycle for delegation

Evidence-Grounded Agent Routing treats routing as a small decision system. Hard constraints are evaluated before optimization, confidence is calibrated, and the router can choose a safe non-selection action when the available evidence is insufficient.

  1. 01RequestIntent and constraints
  2. 02CandidatesRecall before ranking
  3. 03EvidenceSource-linked snapshots
  4. 04EligibilityPolicy and operations
  5. 05DecisionSelect or defer
  6. 06OutcomeEvaluate and recalibrate

The resulting action can be select, parallel, clarify, abstain, or escalate. The architecture avoids treating forced selection as the only successful outcome.

The Routing Decision Record

Every consequential route produces a versioned Routing Decision Record, or RDR. It connects the pre-decision evidence to the eventual outcome and preserves what is needed to replay or challenge the delegation.

EvidenceCandidate facts with sources, timestamps, and freshness
EligibilityCapability, authority, policy, availability, tools, and budget
SelectionRaw fit, calibrated probability, utility, and reason codes
AlternativesRejected candidates and decision-relevant counterfactuals
ContractDelegated scope, resources, limits, and stop conditions
OutcomeObserved result for evaluation, drift detection, and recalibration

What the reference implementation demonstrates

The companion repository implements the control path in Python and LangGraph with typed records, policy and operational gates, configurable scoring, abstention, provenance capture, replay checks, and automated tests.

0.9667full-config route accuracy
0.5667score-only baseline
0.0000measured constraint violations
16automated tests
Claim boundary

These are development results from 42 authored synthetic cases: 12 calibration cases and 30 evaluation cases. They demonstrate that the architecture is executable and that its safeguards affect behavior within the tested scenarios. They are not independent evidence, production validation, or a claim of general superiority. One known evaluation case remains visible in the repository.

The contribution is integration, not invention by renaming

Capability matching, policy engines, confidence calibration, abstention, provenance, and evaluation each have established histories. The paper’s contribution is to integrate them into one operational lifecycle for agent delegation.

The router no longer emits only an agent identifier. It produces a falsifiable decision claim: given this request, candidate set, evidence, policy, and system version, this action was eligible and preferable—and the record exists to examine that conclusion later.

Read, run, inspect, and challenge the work