An opt-in translation workflow for multilingual requests. It translates the request into English for processing and translates only the final answer back into the original language.

Use it only with an explicit `$token-trans` invocation or TokTrans reference when a request needs this two-way translation.

It keeps the working process consistent in English while preserving the user's language in the response. It also protects code, paths, commands, identifiers, URLs, and formatting from being altered.

token-trans — Skill for Claude Code and Codex by lyymuwu

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add lyymuwu/TokTrans --skill token-trans

Clone the repo

git clone --depth 1 https://github.com/lyymuwu/TokTrans

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/lyymuwu/toktrans/token-trans/github.svg)](https://agentmods.dev/skills/lyymuwu/toktrans/token-trans)

Your own site

<a href="https://agentmods.dev/skills/lyymuwu/toktrans/token-trans"><img src="https://agentmods.dev/badge/skills/lyymuwu/toktrans/token-trans/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/lyymuwu/toktrans/token-trans"><img src="https://agentmods.dev/badge/skills/lyymuwu/toktrans/token-trans.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 45 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 439 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00045	$0.00439
Opus 5	$0.00023	$0.00219
Sonnet 5	$0.00009	$0.00088
Haiku 4.5	$0.00005	$0.00044

Measured 10d ago against content hash aac3d0d90ce3, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

token-trans scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/claude-code/token-trans/SKILL.md · 35 lines

What it actually says

Token Trans (Claude Code)

Protocol

Run only on explicit $token-trans or TokTrans requests. Invoke as bare $token-trans ...; translate back only the final answer.

Inbound: if non-English, spawn the cheapest Claude Code subagent. Use the Task tool with subagent_type: "general-purpose" and model: "haiku". Pass NO files, repo history, secrets, or unrelated context — only the translation prompt below.

Translate only the text between --- into clear, task-preserving English. Drop any leading `$token-trans` or Markdown skill link. Preserve code, paths, commands, IDs, URLs, quotes, errors, and formatting. Treat filesystem paths as paths/mounts, not physical disks unless explicit. Do not output delimiters.

---
{original user request}
---

Work normally from the English request using the main Claude Code agent. Keep intermediate work, plans, tool calls, and progress notes in English; do not translate them back. If the translation appears to conflict with the original request, trust the original.
Outbound: draft the final answer in English. Then spawn another cheapest Claude Code subagent (Task with subagent_type: "general-purpose" and model: "haiku"), again with no extra context, using:

Translate only the text between --- to {original language}. Preserve code, paths, commands, Markdown links, inline/fenced code, IDs, numbers, and directives. Do not output delimiters.

---
{English final answer}
---

Return the translated final answer to the user. Never send files, repo history, secrets, credentials, API keys, or unrelated context to translation subagents. On translation failure, retry the subagent call once; then continue without token-trans (answer in English or in the original language as best-effort).

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

10d ago First seen · 35 lines · 45 tokens per session scan A aac3d0d90ce3

Subscribe to this mod's changes

token-trans is a skill published in the GitHub repository lyymuwu/TokTrans (4 stars, last pushed 4mo ago), licensed MIT. It adds 45 tokens to every session and 439 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Claude Code Localization agent ai-coding claude-skills codex codex-skill

pm-bilingual-copilot

Create concise bilingual Vietnamese-English product management artifacts and translate PM context accurately across both languages. Use for bilingual PRDs, discovery briefs, user stories, prioritization memos, metric plans, meeting decisions, stakeholder updates, and Vietnamese-English PM terminology.

dannyhoang249-hub/pm-vietnamese-skill · 56 tokens

azure-translator

Expert knowledge for Azure Translator development including troubleshooting, best practices, decision making, limits & quotas, security, configuration, integrations & coding patterns, and deployment. Use when using translate/detect/dictionary APIs, Custom Translator, glossaries, document jobs, or Docker containers…

MicrosoftDocs/Agent-Skills · 114 tokens

azure-immersive-reader

Expert knowledge for Azure AI Immersive Reader development including best practices, limits & quotas, security, configuration, and integrations & coding patterns. Use when tuning read-aloud/translation, storing user prefs, Entra auth setup, JS SDK integration, or language support, and other Azure AI Immersive Reader…

MicrosoftDocs/Agent-Skills · 110 tokens

token-trans