More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Humanities Book Translation

    1

    Translate scholarly texts between your chosen languages, with contextual terminology, bilingual alignment and traceable review.

    Free

    1 useSecurity scanned
    Humanities Book Translation

    Humanities Book Translation

    Example session with this skill installed

    Translate this selected French sociology passage into scholarly, readable English. Preserve the author’s qualifications and conceptual distinctions. Use the article context to distinguish redress from financial compensation, keep normativity as an analytical concept, and briefly explain the terminology. Translate only the three sentences below, not the whole article.

    French source
    Dans ce contexte, il s’agissait pour nous de relancer l’enquête sur la diversité et la complexité des attentes de réparation que des personnes forment elles-mêmes face aux situations auxquelles elles sont confrontées. Nous voulions prendre au sérieux la normativité des victimes, non pour la défendre, mais pour mieux la comprendre. Nous cherchions à interroger, sous cet angle, la spécificité des collectifs qu’elles forment, et à saisir la force des dispositifs qu’elles rencontrent, qu’elles mobilisent, ou qu’elles créent pour faire valoir leurs attentes de réparation.

    • Read your context and instructions
    • Compiled the humanities book translation

    English translation

    Against this background, we sought to reopen inquiry into the diversity and complexity of the expectations of redress that people themselves develop in response to the situations they face. We wanted to take victims’ normativity seriously, not to defend it, but to understand it better. From this perspective, we sought to examine what is distinctive about the collectives they form and to grasp the power of the arrangements they encounter, mobilize or create to assert their expectations of redress.

    Translation choices

    • “Réparation” becomes “redress”: the authors’ preceding definition includes truth, punishment, apologies, compensation and care, rather than financial compensation alone.
    • “Non pour la défendre” keeps the distinction between understanding victims’ normativity and endorsing it. “Normativity” retains the authors’ analytical concept.
    • “Dispositifs” becomes “arrangements”: the next section describes material and linguistic configurations, including trials, funds and media; translating it as software devices would misidentify the concept.

    Source and license

    Janine Barbot and Nicolas Dodier (2023), Construire une approche sociologique de la réparation, Introduction, printed p.
    4. This demonstration selects the first three complete sentences of the paragraph beginning “Dans ce contexte”; the final sentence is outside the selected scope. Original PDF line breaks and indentation were removed, and the emphasis on “elles-mêmes” is recorded separately.

    Source article · DOI

    Copyright © 2023 Janine Barbot and Nicolas Dodier. Licensed under Creative Commons Attribution 4.0 International.

    English translation/adaptation prepared for this product demonstration using the Humanities Book Translation Skill workflow. The source authors and journal do not endorse the product. This is a short agent-authored translation with agent review; no human specialist has approved it, and it is not a comparative benchmark or an evaluation of a complete book.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Translate humanities, history and social-science books, articles or selected passages with a record you can check. Choose the source and target languages, the audience and the scope. Your AI agent writes the translation; this skill gives it an editorial workflow and document tools.

    Keep the original source, align it with the draft, record terminology in context and review the argument sentence by sentence. Separate checks cover source accuracy, fidelity, target-language prose and consistency. Reviews are tied to the exact source and draft, so changes can make earlier checks stale.

    Start with editable Markdown, aligned JSONL and a bilingual HTML review page. Supported EPUB projects can be rebuilt. Optional local tools add PDF extraction, scanned-page OCR, DOCX input and ordinary DOCX/PDF manuscript output. These exports are reading manuscripts, not automatic reconstructions of every source layout.

    On Agensi, supporting files load on demand through the agent connection. The included session runner executes the Python helpers in memory. Manuscripts, project records and finished outputs can be saved; the skill source stays in the session.

    You need a compatible agent host, its model and Python 3.10+ for the tools. Model usage and optional local dependencies are separate. Translation quality depends on the model, languages, subject and source. Agent review does not replace a qualified human review when the work requires one.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Translation & Localization.

    Source and target languages are selectable; quality depends on the chosen agent model, subject and source. Model usage is not included. Optional local OCR needs Tesseract, language packs and source-image checking. Ordinary DOCX/PDF exports are reading manuscripts: arbitrary page layouts, images, native footnotes and complex-script shaping may require other tools. Supported EPUB rebuilding still needs reader-level visual checks. Agent review is not human specialist approval. The public examples are selected excerpts, not a comparative benchmark or whole-book quality study.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 26 days ago

    • Free with an Agensi account

    Software
    Python 3.10+
    Optional
    TesseractPyMuPDFPillowpython-docxReportLab

    The agent must be able to run Python code. Install Tesseract with the language packs you need.

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access
    Environment Variables

    Reads authorized source files and saves translation projects, review records and deliverables. Python helpers execute in the agent session without saving Skill source. Network access retrieves Agensi supporting files and, when authorized, online sources or optional dependencies. Local OCR does not send manuscript pages to a cloud OCR API; the selected agent host processes text through its own model service. Environment variables are used only for local runtime settings such as TESSDATA_PREFIX. No email, payment, calendar, contact or translation API credentials are required.

    Listed26 days ago

    What's inside

    Frequently Asked Questions

    Specify the source and target languages, including a regional variety when needed. English and Chinese are not default targets. Your agent supplies the language ability, so use a capable model and seek specialist review for difficult material.

    No. The current agent model writes and reviews the text. The Python tools prepare sources, track reviews and export files; they do not call a translation API. You use your existing host/model and pay its normal usage costs.

    It can prepare scanned pages with optional local Tesseract, the required language packs and PyMuPDF. Inspect the recognized text against the page images before translating. OCR confidence alone is not a source-quality check.

    Supported EPUB projects can retain and rebuild their structure. DOCX/PDF output is an ordinary reading manuscript, not a facsimile of an arbitrary original. Complex objects, images, native footnotes, fonts and scripts may need separate preparation or another renderer.

    The workflow records source, calibration, fidelity, style and consistency checks for the selected scope. A release check verifies the records and mechanical conditions. It does not prove semantic perfection, expert approval or publication acceptance.

    New in Translation & Localization

    Free