Vega research copilot

Run run-vg-2

Done with warnings

Started 2026-08-27 · three stages — A to C already happened on the connectors

Read the report

Stages complete

3 / 3

Spend vs cap

$14.90 / $17.86

Elapsed

17m 31s

Coverage

58%

Recommendation · Replay · Cost report

Each stage shows what it was handed as well as what it produced. Ringed steps need a person.

  1. Input

    Workloads in scope
    7
    Snapshots
    snap-2026-08-27-9d04
    Residency
    no constraint

    Output

    Recommendations
    7MODELLED
    Withheld
    1 — evidence truncated

    • R1.1Propose target model and configurationAUTO

    • R1.2Check target availability in the pinned regionAUTO

    • R1.3Withhold where evidence is truncatedAUTO

    • Where truncation corrupted the evidence, the recommendation was withheld rather than guessed.
run outputarchived
[R1]run run-vg-2  assessment=as-vega  started=2026-08-27
[R1][R1.1] proposing targets for 7 workloads
[R1][R1.2] region check against no constraint
[R1][R1.3] 1 recommendation withheld — evidence truncated
[R2][GATE.approved] replay released
[R2][R2.2] replayed 1200 calls · spend $14.90 of $17.86
[R2][R2.3] judge verdict: HOLDS
[R3][R3.1] priced today's configuration from the billing export
[R3][R3.2] scenarios S1–S3 built with ranges
[R3][WARN] coverage 58% — a large share of the bill is unattributed
[R3][done] provenance signed

Completed in 17m 31s · replay spend $14.90 of $17.86 approved MEASURED · saving $1,063/month MODELLED