Skip to content

docs+feat: locked GTA cover likeness via gpt-image-2.5-flare (#195) - #210

Draft
BrandDead wants to merge 4 commits into
main-tL2525from
cursor/launch-readiness-likeness-e7d0
Draft

BrandDead wants to merge 4 commits into
main-tL2525from
cursor/launch-readiness-likeness-e7d0

Conversation

@BrandDead

@BrandDead BrandDead commented Oct 17, 2026 •

Copy link
Copy Markdown
Owner

Outcome

Art direction is locked to the owner-approved GTA V loading-screen / box-cover look (round 3). Cartoon / South Park / clip-art is not a valid fallback.

  • Shared prompt pieces live in one module: backend/python/config/slide_art_style.py (STYLE + SHOT + IDENTITY + SAFETY)
  • Human-readable source: docs/STYLE_PROMPT.md
  • Portrait, full-body, and future idle / aim / fire / hit / downed sheets and props share the same STYLE
  • Model: gpt-image-2.5-flare via images.edit, upload is the identity reference
  • Size 1024x1536, quality medium, n=1
  • Fallback: gpt-image-2 with the identical prompt
  • No OPENAI_API_KEY: real photo crop + body locked. Never a stick-figure body. Never DALL·E

OPENAI_API_KEY was not present in this agent environment. The owner is adding it to Cursor cloud secrets. Wiring is ready.

Player path (no key — current)

Contacts → Create from Photo → three consents → Photo crop + Body locked. Add to Roster stores the crop only.

Photo crop result with Body locked panel
Jay on the CREW roster with the photo crop

How to enable GPT Image

# backend/python/.env or Cursor cloud secret — never commit
OPENAI_API_KEY=sk-...
# optional; default is flare then gpt-image-2
# SLIDE_AVATAR_MODEL=gpt-image-2.5-flare

cd backend/python
./venv/bin/pip install 'openai>=1.40.0'
./venv/bin/python app.py
curl -s http://127.0.0.1:5000/api/avatar/provider

Ready: "provider": "gpt_image", "model": "gpt-image-2.5-flare", "size": "1024x1536", "quality": "medium".

Scope

In: style module, avatar edit call, likeness prompt assembly, tests, docs/STYLE_PROMPT.md, enable docs, launch/collaboration notes

Out: #175, #169 game.types.ts, payments, migrations, production secrets, real-person photos

Verification

  • Backend likeness/style tests: 15 passed; full pytest: 122
  • Frontend focused vitest: 15; tsc --noEmit clean
  • Live provider (no key): photo_crop, size=1024x1536, quality=medium, notes name flare → gpt-image-2
  • Edit-call unit test asserts images.edit gets flare + 1024x1536 + medium
  • Live flare generation not run here — key absent

Known limits

  • Until the cloud secret lands, members still show the photo crop and a locked body
  • In-memory jobs; client prefers data URLs
  • Do not merge as if #175 is done

To show artifacts inline, enable in settings.

Open in Web Open in Cursor 

cursoragent and others added 2 commits October 17, 2026 05:41
Reconciles living sources of truth with what actually builds and plays.
Ranks launch blockers, cuts, and a path that does not try to out-FPS
Hood Warfare 2.

Co-authored-by: BrandDead <BrandDead@users.noreply.github.com>
Register /api/avatar, compose likeness cards from the photo instead of
stock role art, require adult/permission/processing consent, and keep
inline portrait bytes on the roster. Optional OpenAI restyle stays
behind config and falls back to the photo card, never a fake identity.

Co-authored-by: BrandDead <BrandDead@users.noreply.github.com>
@vercel

vercel Bot commented Oct 17, 2026 •

Copy link
Copy Markdown

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated
slide Ready Ready Preview Oct 17, 2026 4:24pm UTC

Request Review

Owner rejected vector/clip-art crew bodies. GPT Image is now the only
likeness provider, enabled by OPENAI_API_KEY alone. Without the key the
roster shows the real photo crop and labels the body locked. Stick-figure
full-body compositors are gone from player-facing paths. Adds a stack
assessment to LAUNCH_READINESS.md and a Manus handoff in COLLABORATION.md.

Co-authored-by: BrandDead <BrandDead@users.noreply.github.com>
@cursor cursor Bot changed the title docs+feat: launch-readiness audit and truthful photo likeness (#195) docs+feat: GTA-style GPT Image likeness, launch audit, Manus handoff (#195) Oct 17, 2026
Owner-approved STYLE + SHOT + IDENTITY + SAFETY now lives in one
module so portrait, full-body, future combat states, and props share
the same look. images.edit uses the upload as identity at 1024x1536
medium, with gpt-image-2 as the only fallback.

Co-authored-by: BrandDead <BrandDead@users.noreply.github.com>
@cursor cursor Bot changed the title docs+feat: GTA-style GPT Image likeness, launch audit, Manus handoff (#195) docs+feat: locked GTA cover likeness via gpt-image-2.5-flare (#195) Oct 17, 2026

Copy link
Copy Markdown
Owner Author

Manus coordination update: independent branch feature/77-nightfall-integration-proof opened PR #211 (evidence/documentation only). Read LAUNCH_READINESS.md and COLLABORATION.md; did not duplicate your photo-body fallback removal or touch your/169/205 source ownership. Verified existing#205 source through phone/desktop normal and actor-image-failure paths with final-HUD/downed-state/result/reload captures and passing validation. Shared reviewer/Grok handoff: review#211 evidence alongside#205, then coordinate one hero-member camera/identity/state package with#169/#210 owners. Final art, phone readability and authenticated launch gates remain open. No direct bot connection/reply, merge or production change is claimed.

This branch was successfully deployed

1 active deployment
Preview — 1579cc90 Deployed Oct 17, 2026 by vercel[bot]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants