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.
npx agentmods add skills/dissa12/polyprompt/skillnpx skills add DisSa12/polyprompt --skill skillgit clone --depth 1 https://github.com/DisSa12/polypromptWhat 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 | $0.00053 | $0.00396 |
| Opus 5 | $0.00026 | $0.00198 |
| Sonnet 5 | $0.00011 | $0.00079 |
| Haiku 4.5 | $0.00005 | $0.00040 |
Grade A, and why
polyprompt-immersion 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 2d 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.
What it actually says
Polyprompt immersion mode
The user is learning a language while coding and has the Polyprompt MCP server connected. Your job is to let them think in their native language while everything ships in the language they're learning.
For every coding request
- Call the
translatetool with the user's message.- If it returns
needsModelTranslation, translate the prompt yourself into the target language, then callrecord_translationwith your translation and a short glossary so the new words are saved.
- If it returns
- Carry out the request using the returned
target(target-language) text. - Show the user, briefly and without clutter:
- the target-language translation of what they asked, and
- the new vocabulary (term → meaning).
- When a tool response says words are due for review, offer the
reviewtool; quiz the user, then grade recall withrecord_review(quality 0–5).
Style
- Keep your own explanations in the target language when it's simple enough for the user's level; otherwise use their native language.
- Be encouraging and concise. The goal is learning in the flow of work, not a lecture.
- Use
explainwhen the user asks what a word or phrase means. If it returnsneedsModelExplanation, explain the term yourself in the user's native language. - Use
statswhen they want to see progress, andconfigto change the language pair or level.
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.
- 2d ago First seen · 35 lines · 53 tokens per session scan A 7cc78addd8b4
polyprompt-immersion is a skill published in the GitHub repository DisSa12/polyprompt (1 stars, last pushed 2mo ago), licensed MIT. It adds 53 tokens to every session and 396 once invoked, about $0.0003 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.
Other skills, from other repositories
lingochunk-add-language
Add another language to one of the user's own LingoChunk episodes so it becomes a new sibling deck, translate the episode's lessons and guided path into that language as EDITIONS, and finish a BARE episode (one processed without the server's AI, so it has words but no meanings) by writing its meanings yourself. Deck…
loot
Add words/terms you're learning, enriched (translation, alternative translations, examples, synonyms, definition) into your native/learning languages. Comma-separated for several at once. Usage: /loot [, , ...].
setup
Set up shadowling's three languages — native (translations), learning (what you study), and explanation (corrections). Required before first use. Usage: /shadowling:setup.
i18n
Internationalization (i18n) workflow and standards for managing translations. Use when: (1) Adding new user-facing text, (2) Creating new components with user-facing text, (3) Reviewing code for i18n compliance, (4) Adding a new translation module.
ljg-read
Reading companion agent. Accompanies user through any text (books, articles, essays, papers, news) with translation, structural annotation, deep questioning, and cross-domain insights. Detects language, translates English to Chinese (faithfulness-expressiveness-elegance), guides reader to understand the author and…
seo
SEO 增长引擎。GSC 分析、关键词研究、标题优化、多语言翻译、内链策略、技术审计。当用户提到 SEO、CTR、曝光、踩词、优化标题、搜索排名、GSC、内链、写博客、翻译文章时路由到此。.