Skip to main content
A multi-agent run is only useful if you can trust the result. Voss persists a deterministic, replayable trail of every run, so you can review it like a PR — and tell what the EM claimed apart from what was independently verified. The trail is built from three durable artifacts, all under .voss/sessions/<root_id>/.

​
What persists

ArtifactWhereContents
Run recordRunFinal (per run)Goal, card counts, done/blocked/killed/rescope counts, EM iterations
Review sidecars<node_id>.review.jsonReviewer-A verification + Reviewer-B verdict + final outcome
Session-tree nodes<node_id>.jsonBudget envelope, scope, role, transitions, terminal state, rejected raises

​
Reading a run

voss board <root_id>          # cards and their final columns
voss review <root_id>         # Reviewer-A + Reviewer-B per card
voss session tree <root_id>   # per-node budget, scope, status
voss session tree <root_id> --json
Together these answer:
  • What changed — transitions and outcomes per card, from the board and session tree.
  • Claims vs. evidence — the EM’s routing rationale and RunFinal summary sit beside the independent reviewer verdicts and evidence references. Unsupported claims are detectable because the verdict carries evidence_refs.
  • Why it passed — or didn’t — Reviewer-A’s verification and Reviewer-B’s verdict are recorded separately in each sidecar.
  • What’s still risky — blocked, killed, and rescoped lineage stays inspectable; budget-raise attempts are recorded as rejected_raises.

​
Determinism

The board and session-tree views are rendered from persisted node files, not from the live agents — so a run reconstructs identically after the fact, and the same --json export drives an ADE audit view rendered as a navigable session tree.
A single unified voss audit <run_id> report and the navigable ADE audit panel are on the roadmap. Today the same evidence is available through voss board, voss review, and voss session tree.