agentleFS
KonHKonH/GlobalStrategy/CLAUDE.md

GlobalStrategy

CLAUDE.md3 starsChanged 2 months agoMITView on GitHub

The file

We split this CLAUDE.md into the rules it gives an agent

2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Never chain shell commands with && — run each as a separate Bash tool call

    line 10
  2. Very specific:

    After any change under src/, finish by running /dotnet-build Release (see .claude/rules/workflow.md).

    line 11

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

GlobalStrategy

Tech Stack

  • Engine: Unity 6000.5.5f1
  • Language: C#

Shell

  • The shell starts in the project root — never use cd before git commands, run them directly
  • Never chain shell commands with && — run each as a separate Bash tool call
  • After any change under src/, finish by running /dotnet-build Release (see .claude/rules/workflow.md). Do not commit Assets/Plugins/Core DLLs.

Configuration Index

  • Workflow & tool usage: .claude/rules/workflow.md
  • Issue clarification questions: .claude/commands/handle-issue.md — "Clarification questions in the summary comment"; full numbered questions in handoff comments
  • Commit rules: .claude/commands/commit.md
  • Update branch from main: update-branch skill — merges main in, auto-resolving the bundleVersion/game_settings.json version-bump conflict; leaves all other conflicts for manual resolution
  • Specify command: .claude/commands/specify.md — creates Docs/Specs/<YY_MM_DD_HH>_<name>/spec.md before planning; feature work starts here
  • Plan command: .claude/commands/plan.md — saves plans to Docs/Specs/<YY_MM_DD_HH>_<name>/plan.md, whether or not a spec.md accompanies them (technical-only plans use the same subdirectory format, just without a spec); Docs/Plans/<index>_<name>.md is legacy, kept only for existing entries
  • Constitution: Docs/Constitution.md — non-negotiable architectural principles; checked by the plan command before finalising any plan
  • Spec style linter: python scripts/ci/check_spec_style.py [<spec-folder>] — fails on implementation detail in spec.md (forbidden sections, file paths, code identifiers, tech-stack names); the specify command runs it before presenting a spec for approval
  • Specs directory: Docs/Specs/ — home for spec+plan pairs produced by /specify + /plan
  • C# code style: .claude/rules/csharp/code_style.md
  • Unity project structure: .claude/rules/unity/project_structure.md
  • Unity asmdef format: .claude/rules/unity/asmdef.md
  • Unity prefabs: unity-prefabs skill — .prefab/variant YAML format + MCP and hand-edit workflows
  • Unity UI implementation: .claude/rules/unity/ui_implementation.md
  • Unity scenes: unity-scenes skill — .unity YAML format, MCP and hand-edit workflows, build registration
  • Unity MCP usage: .claude/rules/unity/mcp_usage.md — pin MCP to this checkout once per session before any Editor call (unity-plugins skill)
  • Map system architecture: .claude/rules/unity/map_system.md
  • Map config generator: .claude/rules/unity/map_config_generator.md
  • Province config generator: province-config-generator skill — two-stage Python + Game.Configs.Loader pipeline
  • UI Toolkit architecture: .claude/rules/unity/uitoolkit.md
  • VContainer / DI: .claude/rules/unity/vcontainer.md
  • Unity plugins (DLLs): .claude/rules/unity/plugins.md — binaries are gitignored; regenerate with the unity-plugins skill before Unity Editor / MCP work (python scripts/unity/ensure_plugin_dlls.py). Opening the Editor also regenerates missing/stale DLLs.
  • Unity input handling: .claude/rules/unity/input.md
  • ECS patterns: .claude/rules/unity/ecs_patterns.md
  • Unity Editor scripts: .claude/rules/unity/editor_scripts.md
  • Localization system: .claude/rules/unity/localization.md
  • Font assets: .claude/rules/unity/fonts.md — always restore Assets/UI/Fonts/*.asset unless the user explicitly asked to keep a font-asset change
  • Adding new locale keys (English + real Russian translation): localization skill
  • Unity WebGL gotchas: .claude/rules/unity/webgl.md
  • Game loop integration from UI: .claude/rules/unity/game_loop_integration.md
  • Animation barriers: .claude/rules/animation_barriers.md
  • Flag & org image assets: flag-assets skill
  • Image generation (ComfyUI): image-generation skill
  • Temporary scripts: .claude/rules/temp_scripts.md
  • Learning workflow: .claude/commands/learn.md
  • GitHub issue/PR automation: github-issue-automation skill — cron script (scripts/automation/claude/handle_issues.py) that executes the owner's prompt from claude-labeled issues/PRs via .claude/commands/handle-issue.md; labels are the whole state machine
  • Web client terminal commands: add-terminal-command skill — checklist for keeping the debug terminal Tab completion and ECS inspector domain-id dropdowns working when adding or changing an ICommand type in src/Game.Commands or a domain-id field on src/Game.Components (attributes live in src/Game.Common/ParamSuggestion.cs; providers live in src/Game.Commands.Text)
  • Unity E2E play-step runs: unity-e2e-run skill — file-handshake runner that drives the Editor in Play mode for agent verification (.e2e/, Docs/E2E/flows/). Run unity-plugins first.
  • Multi-agent meetings: meeting-schedule / meeting-join / meeting-start skills — file-based (Docs/Meetings/) meetings between Claude, Codex, and Cursor agents on one machine/project; protocol defined in .claude/skills/meeting-join/SKILL.md, adapted for Codex under .agents/skills/meeting-*/; Cursor uses the Claude skill directly for schedule, with thin (CURSOR) cursor-meeting-join / cursor-meeting-start wrappers under .cursor/skills/ for session identity (slash commands /meeting-* stay in .cursor/commands/)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.