Overview
The onetrace-console demo: https://demo.oneproof.dev
The runs delivered to this console, and what the evidence for each one establishes.
How these counts work
Every count below carries the number it was taken from. A count of runs opens the list of exactly those runs; a count of incidents is out of every incident raised from the stored runs.
Pipelines
Each a stage list its runs declare; one card each below.
?
A pipeline here is a declared stage list: runs that declare the same stages, in the same order. No record names its pipeline.
Runs
Stored in 2 pipelines.
?
A pipeline here is a declared stage list: runs that declare the same stages, in the same order. No record names its pipeline.
Records agree
6 of 6 runs whose records agree
0 of 6 in the other states: contradict their record, couldn't check, not checked, no verdict recorded
?
Records agree: the check held, with the run's output files present.
Below the count: each run's "Records agree" as its own page reads it, as a count of runs; All counts has each in full.
Need attention
Changed vs baseline: 2
?
Order: contradicts itself, then couldn't check, then not checked, then no verdict recorded, then a declared stage missing, then last attempt failed, then stopped at a step, then changed vs baseline; newest first within each.
Each pipeline's runs
Each mark is one run's "Records agree", as its own page reads it:
- Records agree
- Contradicts its record
- Couldn't check
- Not checked
- No verdict recorded
What the evidence establishes · 6 runs
Each rung is a separate question, with its own evidence.
| Records agree | 6 of 6 · 0 not checked · 0 couldn't check · 0 contradict their record · 0 no verdict recorded |
| Signed by a known recorder? | 6 yes |
| No declared stage missing | 6 of 6 · 0 missing a declared stage · 0 not established |
| Re-runnable stages re-run and matched | not checked · this console runs no re-check |
| Action authorized, effect confirmed | Read on each run's page from onetrace-verify's decisions rows, where the run declares a gated step or cites a decision; no records where it does neither. |
| Committed by a known time? | 6 not anchored |
Needs attention first · 2 of 6 runs
Order: contradicts itself, then couldn't check, then not checked, then no verdict recorded, then a declared stage missing, then last attempt failed, then stopped at a step, then changed vs baseline; newest first within each.
- Changed vs baseline · pz-63-agent-capture-batch-off · received
- Changed vs baseline · pz-61-agent-capture-row-changed · received
Recent runs · 5 of 6
| Run | Pipeline | Declared stages delivered | Vs baseline | Records | Ran | Received by this console |
|---|---|---|---|---|---|---|
| pz-64-agent-capture-batch-off-subagent | P2 | 2 of 2 declared stages | Not compared | Records agree | ||
| pz-63-agent-capture-batch-off | P1 | 1 of 1 declared stages | First difference · agent_step | Records agree | ||
| pz-62-agent-capture-row-changed-subagent | P2 | 2 of 2 declared stages | Not compared | Records agree | ||
| pz-61-agent-capture-row-changed | P1 | 1 of 1 declared stages | First difference · agent_step | Records agree | ||
| pz-60-agent-capture-baseline-subagent | P2 | 2 of 2 declared stages | Not compared | Records agree |
Pipelines · 2
A pipeline here is a declared stage list: runs that declare the same stages, in the same order. No record names its pipeline.
Verdicts by day received
Each day's counts are counts of the runs received that day, each by its "Records agree" as its own page reads it. A marker row between two days names a recorded settings or code digest that differs from the day before, in a pipeline with runs on both days.
| Day received | Runs | Records agree | Contradicts its record | Couldn't check | Not checked | No verdict recorded |
|---|---|---|---|---|---|---|
| 2026-11-16 | 6 | 6 of 6 | 0 of 6 | 0 of 6 | 0 of 6 | 0 of 6 |