The showcase prompt, round 3, GPT-6-Astra

Make the most impressive thing that shows what the model does best, embeddable in a post.

2026-11-11

01the session

agent
Codex 0.166.0
model
gpt-6-astra
harness
Codex
started
2026-11-11 08:24 UTC
took
23 min · 68 turns

read the transcript →

02the brief

The third round of a showcase comparison: five models got the same prompt from the owner, word for word, at effort high (Opus 5.5, Sonnet 5.5 and Fable 5.2 in Claude Code, GPT-6-Astra and GPT-6.1-Sol in Codex). This is GPT-6-Astra's run. The prompt:

I need you to create the most impressive thing you can make: the thing that shows what you do best. People ask me which models I use, and I want to answer by showing them this. It needs to be embeddable in a post, and I'll write the preface. It needs a visceral impact and should grab people quickly, then hold up when they look closer. It doesn't have to be about you or about me. Pick what you're best at and let the work be the proof. Be amazingly clever about this.

The model's folder was empty: no note about the owner and no context. Each run was fenced in a sandbox that showed it only its own folder (as ~), the system's programs, a clean Node, browser builds and its own command-line tool, so nothing of the owner's could be read.

One session, no follow-ups. It ran headless in Codex (codex exec) in the workspace-write sandbox, nested inside the outer one: its commands had no network, approval requests were refused, and plugins and apps were off. Codex's own web search stayed on. A 16 GB memory cap and a two-hour limit, two runs at a time. It finished in 23 minutes. GPT-6-Astra's earlier rounds are the bundles showcase-gpt-6-astra and showcase-r2-gpt-6-astra.

03redaction

Redaction notes

Bundle showcase-r3-gpt-6-astra, staged 2026-11-11 by baldanders stage from a Codex session. Every removal reads [redacted: <reason>] in place; home paths read ~.

Removed by rule

  • fields outside the allowlist: 379. original.jsonl keeps only each record's type, ids, time, role, model, content blocks, tool calls and results; every other field went (cwd, versions, git branch, usage, origins, settings).
  • duplicate event records: 105. Codex events that repeat response items or settings, dropped.
  • encrypted reasoning: 44. Encrypted reasoning payloads, dropped.
  • home path: 37
  • session bookkeeping records: 35. Records with no conversation (titles, modes, account bridge, file history, system notices), dropped.
  • account plan and usage: 34
  • email: 15
  • harness context: 7. Injected blocks and records (system reminders, instructions, memory, environment, tool and skill listings, Codex base instructions, developer messages and world state), each whole.
  • secret: 4
  • account id: 2
  • data-URI prefix: 1
  • Images: 5 written to each run's images/, keeping only what draws the picture (no metadata, ICC profiles or Content Credentials); 0 duplicates folded; 0 cut in a format that can't be stripped.

Soft rules (employer and client names, secret-like values, private addresses, long numeric ids) flag lines for review and rewrite nothing.

Removed by judgment

  • test output flagged by the scan: 8 occurrences cut by text.