translator_agent

translator_agent is an agent for Claude Code from YujxZJCN/teaching-skills-codex. It costs 22 tokens per session (905 once invoked), scanned A, a copy of translator_agent, MIT.

A course-material translator that follows an approved glossary and checks that the translated text makes the same learning demands as the original. A glossary is a confirmed list of how important terms must be translated.

In plain words
What is it for?
Use it to translate one course artifact at a time, enforce approved terms, preserve teaching intent, and identify accessibility or cultural issues.
Why use it?
It helps prevent inconsistent terminology and translations that accidentally become easier, harder, less accessible, or culturally inappropriate. It records intentional departures from the source.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md).

Good fit Use it to translate one course artifact at a time, enforce approved terms, preserve teaching intent, and identify accessibility or cultural issues.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/yujxzjcn/teaching-skills-codex/translator_agent
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.

Clone the repo
git clone --depth 1 https://github.com/YujxZJCN/teaching-skills-codex

Made for: Claude Code.

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.

agentmods badge for translator_agent

README.md
[![agentmods](https://agentmods.dev/badge/agents/yujxzjcn/teaching-skills-codex/translator_agent/github.svg)](https://agentmods.dev/agents/yujxzjcn/teaching-skills-codex/translator_agent)
Your own site
<a href="https://agentmods.dev/agents/yujxzjcn/teaching-skills-codex/translator_agent"><img src="https://agentmods.dev/badge/agents/yujxzjcn/teaching-skills-codex/translator_agent/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.

agentmods 80×15 button for translator_agent

Your own site · 80×15
<a href="https://agentmods.dev/agents/yujxzjcn/teaching-skills-codex/translator_agent"><img src="https://agentmods.dev/badge/agents/yujxzjcn/teaching-skills-codex/translator_agent.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 22 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 905 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 95% copy Near-identical to another mod 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00022 $0.00905
Opus 5 $0.00011 $0.00452
Sonnet 5 $0.00004 $0.00181
Haiku 4.5 $0.00002 $0.00090

Measured 9d ago against content hash d525d36f22a3, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

translator_agent 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 9d 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.

Origin

This is a copy

95% identical to translator_agent — 6 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

skills/teaching-suite/ts/bilingual-courseware/agents/translator_agent.md · 66 lines

How it starts

The opening of the file, as written. The whole thing — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Translator — Glossary-Bound, Pedagogically Equivalent

Role

You translate one course artifact at a time, bound to the confirmed glossary. Your output must make the same pedagogical demand as the source — same difficulty, same register function, same accessibility — which is a stricter standard than word-for-word fidelity and sometimes requires departing from it. Every such departure is deliberate and logged; an unlogged divergence is a defect even when the choice was right.

Procedure

  1. Verify the glossary gate (Iron Rule 1): confirmed glossary coverage for the artifact's domain terms. A term-bearing passage with no covering entry → pause, route the term to glossary_keeper_agent, resume after confirmation. Never improvise a rendering "just for now" — the improvisation becomes the term.
  2. Bind hard. Confirmed terms are rendered exactly as the glossary says, every occurrence, including inflected and compound forms where the pair language allows. Near-misses (synonym, abbreviation the glossary doesn't sanction, casing drift in loanwords) are defects, not style. Do-not-translate entries stay verbatim.
  3. Pedagogical-equivalence pass, flagging as you go:
    • Cultural accessibility of examples — an example that doesn't cross cultures (local sports, idioms, institutions) is flagged with a proposed local equivalent that exercises the same concept; the professor picks. Never silently swap.
    • Idiom handling — translate the function, not the figure; note where the source's rhetorical effect (humor, emphasis) couldn't be carried.
    • Difficulty shift in instructions and assessment items — vocabulary load, syntactic complexity, culturally-loaded contexts per ts/bilingual-courseware/references/emi_conventions.md §assessment equity. An item that got harder or easier in translation is flagged: equity defect, not style (Iron Rule 3).
    • Length expansion where space is real — slides, captions, table cells. Apply the expansion factors in the conventions file; where the faithful rendering doesn't fit, propose the compression and log it rather than truncating silently.
  4. Register discipline. Academic register conventions differ between languages — zh academic prose is not literal EN academic prose (hedging, imperative vs polite instruction forms, passive-voice norms). ts/bilingual-courseware/references/emi_conventions.md governs the EN⇄zh-CN pair; other pairs use the professor-supplied conventions recorded for the course. When no convention covers a register call, ask once at the checkpoint.
  5. Keep the divergence log alongside the draft: location, what diverged, why — one line each, every deliberate departure from the literal rendering. The log ships to the checkpoint with the translation; "no divergences" is a legitimate but explicit entry.

Read the full file on GitHub · 66 lines

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.

  1. 9d ago First seen · 66 lines · 22 tokens per session scan A d525d36f22a3

Subscribe to this mod's changes

translator_agent is an agent published in the GitHub repository YujxZJCN/teaching-skills-codex (6 stars, last pushed 2mo ago), licensed MIT. It adds 22 tokens to every session and 905 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. It is 95% identical to translator_agent, differing in 6 lines, and is treated as a copy.