AbdelStark/worldforge

Harness framework to build world model based workflows for physical AI systems.

9 files for Codex, OpenCode and Claude Code: evaluation-benchmarking, optional-runtime-smokes, persistence-state, provider-adapter-development and 5 more — 12,727 tokens loaded in every session.

Open on GitHub abdelstark.github.io

Instructions

worldforge AGENTS.md AGENTS.md A 8,618 tok
worldforge CLAUDE.md CLAUDE.md A 4,109 tok

Skills

evaluation-benchmarking .codex/skills/evaluation-benchmarking/SKILL.md A 53 tok
optional-runtime-smokes .codex/skills/optional-runtime-smokes/SKILL.md A 59 tok
persistence-state .codex/skills/persistence-state/SKILL.md A 82 tok
provider-adapter-development .codex/skills/provider-adapter-development/SKILL.md A 73 tok
public-docs-release .codex/skills/public-docs-release/SKILL.md A 50 tok
testing-validation .codex/skills/testing-validation/SKILL.md A 61 tok
tui-development .codex/skills/tui-development/SKILL.md A 63 tok

These files are AbdelStark/worldforge's own configuration — they tell Codex, OpenCode and Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.