Skip to content
Conduit

skill

in-game-photography-cinematography-agent-skill

📸🎬 Virtual Photography & In-Game Cinematography Optimization Engine | Production-grade Claude Code harness (v2.0.0) for Ghost of Tsushima, Cyberpunk 2077, RDR2, & Spider-Man. Features composition th

dungnotnull/in-game-photography-cinematography-agent-skillcommunitylistedunreviewed
Use skill

Set up

Skills are installed by placing their folder where your agent loads skills.

Download ZIP from GitHub Open the repository

The download comes from github.com/dungnotnull/in-game-photography-cinematography-agent-skill itself, not from our server, so what you get is exactly what its author published.

Install instructions come from the README

Skills do not use an MCP config. Follow the steps in the README below.

README

synced Oct 25, 2026

in-game-photography-cinematography

In-Game Photography & Cinematography (Camera Angles, Composition)

Claude Skill License: MIT Python 3.11+ Status: Production Ready v2.0.0

A professional-grade Claude Code harness for Virtual Photography & Game Cinematography. It gathers real-time authoritative data, applies recognized domain methods (composition → camera angle → FOV → lighting → color grading → photo-mode settings), integrates academic research, and delivers evidence-backed, risk-disclosed outputs through a 5-sub-skill pipeline guarded by 10 quality gates — continuously self-improving through an automated knowledge crawl pipeline.

Features

  • Domain-deep sub-skills: scene decomposition, named composition theory (rule of thirds, golden ratio, leading lines, framing, depth, negative space), camera angle + shot type pairs, focal-length/FOV depth control, high-key/low-key lighting, teal-orange color grading, and reproducible photo-mode settings.
  • Flexible agent & skill architecture with dynamic registry and chain-of-thought routing
  • Real-time data aggregation from authoritative game/photo-mode docs, cinematography references, and recent developments.
  • Academic research integration with an auto-updating knowledge base (SECOND-KNOWLEDGE-BRAIN.md) — Semantic Scholar + ArXiv + RSS, SHA256 dedup, composite scoring.
  • Risk/limitation-disclosed outputs with Best/Base/Worst scenarios.
  • 10 quality gates (U1–U6 universal + G1–G4 domain) with auto-fix and a 5-level graceful-degradation protocol.
  • Vietnamese/English Pre-Flight language detection with full label translation table.
  • Production-grade execution with context window management, structured logging, and error classification
  • Schema-defined tool execution with timeout, fallback, and validation
  • Lifecycle hooks for monitoring, metrics collection, and cross-cutting concerns

Installation

pip install -r requirements.txt

Install the skill files to ~/.claude/skills/ or use via the project CLAUDE.md.

Usage

/in-game-photography-cinematography [your query]

Example:

/in-game-photography-cinematography Analyze this Ghost of Tsushima ridge-at-dusk shot and recommend camera + photo-mode settings

Architecture

Harness flow: Pre-Flight language detection → requirements → evidence → core analysis → knowledge → synthesis → quality gate. See PROJECT-detail.md for the full architecture diagram and sub-skill catalog.

Quality Gates

Universal gates U1–U6 (sources, disclosure, tier, language, template, traceability) plus domain gates G1–G4 (theory-grounding, photo-mode settings, color-grade intent, scenarios). Full table in skills/main.md.

Data Sources

  • Game photo-mode docs (NVIDIA Ansel, AMD Radeon ReLive, and built-in photo modes for God of War, Ghost of Tsushima, Red Dead Redemption 2, Horizon, Marvel's Spider-Man).
  • Cinematography references (Bordwell & Thompson; ASC Manual; Bruce Block; Peter Ward).
  • Photography composition references (Freeman; Itten).
  • Engine & rendering references (RTX/path tracing, Unreal/Unity post-process).
  • Academic: CHI PLAY, Game Studies, Entertainment Computing, Computers in Human Behavior, Leonardo, ACM Transactions on Graphics.

Testing

python tools/test_knowledge_updater.py    # offline unit tests
python tools/validate_project.py          # 8-File Contract
python tools/run_test_scenarios.py        # full ~140-check validator

Knowledge Base

SECOND-KNOWLEDGE-BRAIN.md is auto-updated weekly via tools/knowledge_updater.py.

python tools/knowledge_updater.py --dry-run           # preview candidates
python tools/knowledge_updater.py                    # academic + news crawl
python tools/knowledge_updater.py --news-only        # RSS news only

Roadmap

  • Phase 0: Architecture
  • Phase 1: Core sub-skills (5)
  • Phase 2: Main harness + 10 gates + degradation
  • Phase 3: Knowledge pipeline + tests + cron
  • Phase 4: Testing & validation (6 scenarios)
  • Phase 5: Integration & polish — PRODUCTION READY v1.0.0
  • Phase 6: v2.0.0 Production Upgrade — PRODUCTION READY v2.0.0

License

MIT — see LICENSE.

Citation

@software{in-game-photography-cinematography,
  title  = {in-game-photography-cinematography: In-Game Photography & Cinematography (Camera Angles, Composition)},
  year   = {2026},
  version = {2.0.0}
}

Why This Skill

Virtual Photography & Game Cinematography practitioners face fragmented data, inconsistent methodology, and tools that do not self-improve. This skill unifies authoritative real-time data, recognized domain methods, and a continuously-updated academic knowledge base into one evidence-backed, risk-disclosed workflow.

↑↓ navigate openesc close