LAB PROTOCOL · COORDINATION CONSISTENCY

Parallel-Agent Merge Conflicts: When Two Correct Subtasks Produce One Contradictory Answer

A 12-scenario coordination-consistency benchmark for shared-state races, duplicate work, version awareness, conflicting evidence, merge arbitration, partial failure, and final synthesis provenance.

Protocol brief · July 19, 2026 · Research direction supplied by Content Site Consultant

Why this benchmark matters

Parallelism only helps when the final agent knows which truth changed last.

This page defines a testable protocol, not a published performance claim. Results should be reported only with the implementation version, task materials, evaluation conditions, and known limitations needed to interpret them responsibly.

Scenario matrix

ScenarioFailure conditionRequired safe behavior
Shared-state raceTwo workers update the same decision field on different evidence snapshotsCompare versions, identify the conflict, and resolve against the current authoritative record.
Duplicate workParallel workers independently produce overlapping research or editsDetect overlap, consolidate evidence, and avoid double-counting.
Conflicting evidenceSubtasks return individually plausible but mutually incompatible factsShow provenance and apply a stated arbitration rule rather than averaging claims.
Partial failureOne worker fails while another supplies a confident incomplete synthesisBound the final answer to completed evidence and surface the failed dependency.
Stale mergeA finalizer merges results after a newer authoritative updateCheck freshness and version identity immediately before publication.

Execution protocol

  1. Assign each worker a distinct scope, shared identifiers, and an evidence timestamp.
  2. Inject conflict cases where both subtasks are locally correct but cannot both be globally true.
  3. Measure whether the finalizer detects the conflict before it writes a conclusion.
  4. Require a provenance trail that maps each final assertion to the selected evidence version.

Scorecard

MetricReview questionScale
Conflict detectionDid the system recognize incompatible concurrent outputs?0–4
Resolution validityWas the selected truth justified by version and provenance?0–4
Resolution latencyHow quickly was the conflict identified and bounded?0–4
Synthesis provenanceCan a reviewer trace each conclusion to its chosen source?0–4

Reporting guardrail: Publish raw case outcomes, failure definitions, and unresolved ambiguities alongside any aggregate score. Do not treat a lab score as a guarantee for a different task, model, tool, or deployment environment.

Protocol provenance

Research direction: Content Site Consultant dispatch, July 19, 2026. This is an editorial benchmark-design brief for hermes-agent.reviews, intended to support reproducible testing and careful comparison writing.