// Translates text between any two languages while preserving source format. Use when the user needs to translate plain text, code, or markdown content.
Shows a report of skill invocations and usage frequency.
Picks the right Kubernetes Deployment update strategy (RollingUpdate / Recreate / Blue-Green / Canary) for the situation. Use when configuring a new Deployment, changing rollout config, or deciding how to ship a risky change.
Runs Multica CLI commands by mapping the user's request to the right `multica <command>` invocation. Use when the user asks to do anything with Multica — issues, agents, workspaces, autopilots, skills, squads, or the daemon.
Runs Multica CLI commands by mapping the user's request to the right `multica <command>` invocation. Use when the user asks to do anything with Multica — issues, agents, workspaces, autopilots, skills, squads, or the daemon.
Diagnoses and fixes CI/CD failures from GitHub Actions logs. Use when a build is broken, CI is red, or a pipeline has failed.
Generates written content — blog posts, social posts, emails, and marketing copy — matched to the project's existing voice. Use when the user needs written material.
| name | language-translate |
| description | Translates text between any two languages while preserving source format. Use when the user needs to translate plain text, code, or markdown content. |
| when_to_use | translate, language translate, translate text, translate code, translate markdown, source to target language, auto-detect language |
| user-invocable | true |
| model | haiku |
| disable-model-invocation | false |
| allowed-tools | Bash, Write, Edit, Read |
Translate content from one language to another while keeping the source format intact. The principle: the reader should be able to tell what the original looked like just by looking at the translation — same structure, same formatting, same code.
The translation should change the words but not the shape of the content.
#, ##), emphasis (**bold**, *italic*), links, lists, tables, and code fences stay exactly as they are. Only the human-readable prose inside gets translated.//, #, --, /* */, etc.)"""...""", '''...''')Technical identifiers look like words but are not — translating them breaks the content. Leave these in the source language:
Example 1 — Markdown:
Input:
# Welcome
This is **important** information.
Output (to French):
# Bienvenue
Ceci est une information **importante**.
Example 2 — Python with comments:
Input:
# Calculate the total price
def calculate_total(items):
"""Return the sum of all item prices."""
return sum(item.price for item in items)
Output (to German):
# Den Gesamtpreis berechnen
def calculate_total(items):
"""Gibt die Summe aller Artikelpreise zurück."""
return sum(item.price for item in items)
Notice that calculate_total, items, item.price, and sum are preserved. Only the human-facing comment and docstring change.
Example 3 — Mixed markdown + code:
Input:
# Usage
Call the API like this:
```javascript
// Fetch the user profile
const user = await fetchUser(id);
```
Output (to Japanese):
# 使い方
このように API を呼び出します:
```javascript
// ユーザープロフィールを取得
const user = await fetchUser(id);
```
キャッシュ (cache).The translate-markdown skill delegates to this skill for the actual translation work. This skill is the general-purpose primitive; translate-markdown is a convenience wrapper for the common markdown-file case.