ozan
bf51d89c1c
Explains what an agent harness is (identity, memory, skills, tools, guardrails) and why game dev needs a specialised one. Ties together all existing posts and skills into the harness framing. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2.0 KiB
Tinqs Studio Blog
Engineering blog from Tinqs Studio --- a game development platform built on a Gitea fork with 3D asset preview, LFS-first workflows, AI agent tools, and creative pipelines for game teams.
We're building Tinqs Studio while using it to make our own game --- a survival colony sim in Godot 4. These posts cover what we've learned along the way.
Posts
- How a Small Game Studio Runs on AI Agents (2026-03-06)
- One Binary to Rule Them All: Building a Studio CLI (2026-05-18)
- Why We Forked Gitea and Built Tinqs Studio (2026-05-20)
- Streaming a 12km Archipelago in Godot 4 (2026-05-22)
- AI Art at Scale: Using fal.ai Flux for Game Asset Generation (2026-05-25)
- Tinqs Studio Is an Agent Harness for Game Dev (2026-05-25)
Skills
Reusable AI agent playbooks. Each skill is a markdown file that teaches an AI agent (Cursor, Claude Code, etc.) a specific workflow. Drop them into your .cursor/skills/ directory.
- Image Generation with fal.ai --- Generate game art using Flux models with structured prompts
- Concept Art Pipeline --- End-to-end 2D concept art to 3D model workflow
- Sora 2 Video Generation --- Generate trailer clips and game footage with OpenAI Sora 2
- 3D Model Generation with Tripo --- Text-to-3D and image-to-3D via Tripo Studio
- Blog Authoring --- Write and publish markdown blog posts
What are skills?
Skills are structured markdown files that give AI coding assistants step-by-step procedures for complex workflows. Instead of explaining the same process every session, you write it once and the agent follows it. Think of them as runbooks for AI agents.
License
Content is CC BY 4.0. Use it, adapt it, credit us.