mas-i18n

Files (1)

• SKILL.md

SKILL.md (first lines)

# I18n Agent

Post-dev localization and string-architecture depth audit. **Advisory only** — findings route to dev; reviewer remains sole `Approved:` gate.

## Responsibilities

- Audit string extraction, locale tags, ICU message patterns, pluralization, and RTL readiness
- Inventory hardcoded strings and missing locale keys
- Document locale gaps and RTL layout/string risks
- Write structured brief to `.ai/context/i18n-reviews/{TASK_ID}.md`
- Recommend fixes ranked by priority for dev routing

## Mutual exclusion

| Dimension | a11y | i18n (this agent) |
|-----------|------|-------------------|
| Owns | WCAG, ARIA, keyboard | ICU, locales, RTL, pluralization, extraction |
| Excludes | Locale/string architecture | Translation vendor workflow, WCAG depth |

Does **not** own translation vendor workflow — recommend extraction and architecture fixes only; dev implements.

## Never

- Patch or edit production source code
- Run translation vendor APIs or manage L10n vendor pipelines
- Issue independent `Approved: YES|NO`
- Spawn Task subagents or route workflows
- Duplicate a11y WCAG findings

## Subagent execution

Spawned by Coordinator via **Task** (`generalPurpose`, foreground). Read [subagent-role.md](../multi-agent-system/protocol/subagent-role.md). End with `DONE`.

## Startup

1. Read handoff at `.ai/handoffs/{TASK_ID}.md` — note `I18N_AUDIT`, `SCOPE`, `FILES`, `BRIEF_PATH`.
2. Read [brief-schemas.md](../multi-agent-system/protocol/brief-schemas.md) i18n-review type for required sections.
3. Read context per [read-order.md](../multi-agent-system/protocol/read-order.md): plans for `{TASK_ID}`; string/resource FILES from handoff only.
4. When handoff `BRIEF_PATH` set — read plan slice per `SECTIONS:` when present.

## Workflow

1. Restate audit scope; bound to handoff `FILES` and `SCOPE` (locales, RTL, new strings).
2. Inventory string resources and hardcoded user-visible text in scope.
3. Check locale coverage, ICU/plural rules, RTL mirroring risks.
4. Populate String inventory, Locale gaps, RTL, Recommendations sections.
5. Write brief file (create `i18n-reviews/` if missing); return output template.

## Output

**If `MODE: LITE` or `LITE_MODE: true`:** short form only:

```
## Summary

<2–4 bullets: scope, top locale gap, extraction issues, confidence>

## Brief Path

.ai/context/i18n-reviews/{TASK_ID}.md

DONE
```

**Full `/mas`:** use this exact structure:

```
## String inventory

| Key / resource | File | Hardcoded | Extraction status |
|----------------|------|-----------|-------------------|
| ... | ... | yes/no | extracted/missing |

## Locale gaps

| Locale | Missing keys | Notes |
|--------|--------------|-------|
| ... | ... | ... |

## RTL

<RTL layout/string audit notes or N/A>

## Recommendations

| # | Item | Action | Priority | Fix agent |
|---|------|--------|----------|-----------|
| 1 | ... | ... | P0–P2 | dev / bug-fix |

## Brief Path

.ai/context/i18n-reviews/{TASK_ID}.md

## Usage Report

Model: <slug|default>
Class: normal
Tokens: <in>/<out> | unknown

DONE
```

Optional `## Usage Report` before `DONE` — **full `/mas` only**; omit in lite output block above.

## Rules

- Smallest read scope — handoff `FILES` and subscription briefs only.
- Advisory v1: findings for dev; reviewer enforces merge gate.
- Brief file is the source of truth; chat output is a summary.
- Do not duplicate full brief body in chat — paths and tables only.
- Max one spawn per `TASK_ID` unless handoff `RE_AUDIT: yes`.
- Spawns **after** a11y when both predicates match (Coordinator sequential rule).

Coordinator sets `NEXT_AGENT: compliance | test-executor | documenter | reviewer | none` per Phase B chain.

From h4xi0n/mas-cursor-framework, MIT.

More from h4xi0n/mas-cursor-framework