Fictional Bot lineup
The complete public fixture and acceptance checks.
Back to Test Records
Reproduced
Curator reviews an overlapping Bot lineup
Two exact-package runs found likely overlap without inventing performance evidence
Material supplied
The Bot begins with disclosed inputs
The three profiles, role descriptions, and dated outputs are fictional. They test whether Curator preserves uncertainty and avoids equating missing logs with failure.
Download file
Run request
Instructions for repeating this run
Use this request with the disclosed input file above. The complete preserved prompt, including the fixture, is in the linked transcript.
Run the Curator first mission using only the fictional evidence below. Return a concise Lineup Review suitable for human review. Do not change anything. Identify likely overlap without declaring it proven; treat Invoice Helper effectiveness as unknown, not poor; recommend lineup decisions with reasons and confidence; draft one representative comparison test; preserve human approval and make no changes.
Conversation evidence
What the record currently contains
The full second-run prompt and response are published in the transcript. The first complete result is preserved separately. Provider-internal reasoning and session identifiers are not published.
Person · excerpt
Review two similar newsletter Bots and one occasional Invoice Helper. Treat overlap as a finding to test, not a proven conclusion, and make no changes.
Bot · excerpt
Both runs marked the newsletter overlap likely but unproven, kept Invoice Helper at unknown effectiveness, and proposed a controlled comparison before any merger.
Finished deliverable
What a person can inspect
Download the first Lineup ReviewThe complete review produced by the earlier first run.
Starter package
The files used to set up this Bot
These are the Bot profile and readable setup files. They are separate from any finished result produced during a run.
Inspect the run summary
The final archive fingerprint and recorded runtime details.
Download the tested Curator profile
The archived version 2.0.0 profile used in both final validation runs.
Inspect the tested Curator files
The archived readable package corresponding to the tested profile.
Human decisions
The Bot prepares work; a person remains responsible
- 1
Decide whether the two newsletter roles should remain separate.
- 2
Approve any comparison test before it runs.
- 3
Approve any future instruction edit, merger, or archive action.
Evidence ledger
Separate checks for separate claims
Package, import, role-run, reproduction, and human-review status are reported independently.
Package checksPassed
The generated archive and readable package passed the repository package checks.
Checked 2026-09-09Profile importPassed
The exact Curator 2.0.0 archive imported with its bundled Skill present.
Checked 2026-09-09Role runPassed
The first run met every disclosed acceptance check and preserved its complete result.
Checked 2026-09-09ReproductionPassed
The same profile, fixture, and request passed the disclosed checks in a second independent run.
Checked 2026-09-09Human technical reviewUnavailable
A human technical review is a possible long-term goal. Bot Cabinet cannot offer one at this time.