fullstack-starter / rules
Cursor rule223 starsChanged 28 days agoMITView on GitHub
Response language rules for agents and workflows
We split this i18n-guide.mdc into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 28
When translating user-facing content, the translation_voice field in .agents/oma-config.yaml controls global rhythm and formality.
- Distinctive: line 73
All workflows follow these rules.
≈ “Common” is an estimate: at least 1 in 5 of the 6,350 Cursor rule 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.
Files
--- description: "Response language rules for agents and workflows" globs: alwaysApply: true --- <!-- Generated by oh-my-agent. Source: .agents/rules/i18n-guide.md --> # i18n Guide — Response Language Rules Rules for determining and applying response language across agents and workflows. ## Language Resolution Response language is determined by the following priority: 1. **User prompt language** — if the user writes in a specific language, respond in that language 2. **`oma-config.yaml`** — `language` field in `.agents/oma-config.yaml` 3. **Fallback** — English (en) if neither of the above is set ```yaml # .agents/oma-config.yaml language: ko # ko, en, ja, zh, ... ``` ## Translation Voice When translating user-facing content, the `translation_voice` field in `.agents/oma-config.yaml` controls global rhythm and formality. It is applied on top of `oma-translation` content-type persona routing. | Value | Effect | |---|---| | `formal` | strict complete sentences, no fragments, formal register only | | `balanced` (default) | content-type defaults — fragments only in label/cell positions | | `interpreter` | punchy, audience-first, spoken cadence; fragments allowed when natural | Workflows that translate user-facing content should respect this setting via the `oma-translation` skill rather than hardcoding a tone. ## What to Localize | Category | Localize? | Example | |----------|-----------|---------| | Natural language responses | Yes | User-facing explanations and descriptions | | Error messages (user-facing) | Yes | Authentication failure messages | | Status updates / progress reports | Yes | "Phase 2 complete, starting Phase 3" | | Charter Preflight output | Yes | Descriptive text localized, keywords stay in English | | Result files (result-*.md) | Yes | User-readable text | ## What Stays in English | Category | Why | Example | |----------|-----|---------| | Code (variables, functions, classes) | Codebase consistency | `getUserProfile()` | | Git commit messages | Conventional commits standard | `feat: add user auth` | | PR titles/body | GitHub collaboration standard | `fix: resolve race condition` | | Technical / domain terms | Meaning loss when translated | API, JWT, middleware, scaffold | | File paths / config keys | System identifiers | `.agents/config/` | | Log levels / status keywords | Parsing compatibility | `Status: completed`, `BLOCKED` | | CLAUDE.md / SKILL.md content | System prompts | Keep English originals | ## Mixed-Language Rules 1. **Keep technical terms in original language** — don't force-translate - Good: "Validates the JWT token" - Bad: "Validates the JSON Web Token" 2. **Code blocks are always in English** — including comments 3. **Inline code (`backtick`) is never translated** 4. **Parenthetical supplement allowed** — for unfamiliar terms, use `translated(original)` format once 5. **Register consistency** — match the target language's appropriate register for the context 6. **Translation tasks** — for translating UI strings, docs, or marketing copy, use the `/oma-translation` skill ## Workflow Integration All workflows follow these rules. The following line in existing workflows references this guide: ``` - **Response language follows `language` setting in `.agents/oma-config.yaml` if configured.** ``` > ARB-based localization (ARB as the single cross-platform source of truth for > UI strings) is a path-scoped rule — see `i18n-arb.md`, which loads only when > `*.arb` files are in context. ## Subagent Behavior - Subagent **result files** (`result-*.md`) are written in the user's language - **Internal communication** between subagents (charter, status keywords) stays in English - Agent definition files (`.agents/agents/*.md`) stay in English
Where files are headingNew Cursor rule files have more rules about planning and tracking tasks, and fewer about code style.
Of the Cursor rule files we read, the 3,978 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository131 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .cursor/rules/backend.mdcCursor rule
- .cursor/rules/branching-strategy.mdcCursor rule
- .cursor/rules/build-guide.mdcCursor rule
- .cursor/rules/commit.mdcCursor rule
- .cursor/rules/database.mdcCursor rule
- .cursor/rules/debug.mdcCursor rule
- .cursor/rules/design.mdcCursor rule
- .cursor/rules/design-tokens-guide.mdcCursor rule
- .cursor/rules/dev-workflow.mdcCursor rule
- .cursor/rules/frontend.mdcCursor rule
- .cursor/rules/i18n-arb.mdcCursor rule
- .cursor/rules/infrastructure.mdcCursor rule
- .cursor/rules/lint-format-guide.mdcCursor rule
- .cursor/rules/market.mdcCursor rule
- .cursor/rules/mobile.mdcCursor rule
- .cursor/rules/preferred-editing-tools.mdcCursor rule
- .cursor/rules/quality.mdcCursor rule
- .cursor/rules/test-guide.mdcCursor rule
- oma-academic-writingSkill
- oma-architectureSkill
- oma-backendSkill
- oma-brainstormSkill
- oma-coordinationSkill
- oma-dbSkill
- oma-debugSkill
- oma-deepsecSkill
- oma-designSkill
- oma-dev-workflowSkill
- oma-docsSkill
- oma-explanationSkill
- oma-frontendSkill
- oma-hwpSkill
- oma-imageSkill
- oma-marketSkill
- oma-mobileSkill
- oma-observabilitySkill
- oma-orchestrationSkill
- oma-pdfSkill
- oma-pmSkill
- oma-qaSkill
- oma-recapSkill
- oma-refactorSkill
- oma-scholarSkill
- oma-scmSkill
- oma-searchSkill
- oma-skill-creationSkill
- oma-slideSkill
- oma-tf-infraSkill
- oma-translationSkill
- oma-videoSkill
- oma-voiceSkill
- architectureSkill
- brainstormSkill
- commitSkill
- convertSkill
- coordinate-proSkill
- coordinateSkill
- debugSkill
- deepinitSkill
- The first 60 shown.
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
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.

