A local document-scrubbing tool that replaces personal details with tags before text is sent to a language model, then restores the real details in the returned answer.

It helps prepare contracts and customer exports for model processing, check for personal data, create masked copies, and restore the original values locally.

It keeps names, phone numbers, addresses, and other personal data from leaving your computer while still letting the model process the document.

docs-masked — Skill for Claude Code by kpshinnik/docs

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 kpshinnik/docs_masked --skill docs-masked

Clone the repo

git clone --depth 1 https://github.com/kpshinnik/docs_masked

Made for: Claude Code.

Or install docs-masked, the plugin that ships this one along with the rest of its 1 skill.

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/kpshinnik/docs_masked/docs-masked/github.svg)](https://agentmods.dev/skills/kpshinnik/docs_masked/docs-masked)

Your own site

<a href="https://agentmods.dev/skills/kpshinnik/docs_masked/docs-masked"><img src="https://agentmods.dev/badge/skills/kpshinnik/docs_masked/docs-masked/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/kpshinnik/docs_masked/docs-masked"><img src="https://agentmods.dev/badge/skills/kpshinnik/docs_masked/docs-masked.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 296 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,714 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. Third-party audits

NVIDIA SkillSpector warn 7 Sept 2026

SkillSpector: 1 finding, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

medium analysis-evasion · line 1
Suspicious Unicode normalization or mixed-script content

Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.

How audits are shown

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	$0.00296	$0.01714
Opus 5.5	$0.00118	$0.00686
Sonnet 5.5	$0.00059	$0.00343
Haiku 4.5	$0.00030	$0.00171

Measured 1mo ago against content hash 527999ef097c, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

docs-masked 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 1mo 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/docs-masked/SKILL.md · 159 lines

How it starts

The opening of the file, as written. The whole thing — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.

docs-masked — обезличивание документов до отправки в модель

Зачем

Когда в модель уходит договор или выгрузка клиентов, туда уходят и чужие персональные данные. Скилл превращает это в контролируемую операцию: данные заменяются тегами на локальной машине, наружу уходит текст без людей, а ответ модели восстанавливается обратно по сейфу, который тоже никуда не уезжает.

документ ──▶ маска ──▶ контроль утечки ──▶ модель ──▶ обратная подстановка
           локально      локально          сеть           локально

Порядок работы

Он всегда один и тот же — не пропускай шаги, даже если документ выглядит безобидно. Пропуск второго шага и есть та ситуация, ради которой скилл нужен.

1. Посмотреть, что внутри. Локально, без сети:

python3 -m pii_shield.cli scan договор.docx

Вывод показывает типы и количество находок. Если ожидал персональные данные, а находок нет — это повод разобраться, а не идти дальше.

2. Обезличить и убедиться, что чисто.

python3 -m pii_shield.cli mask договор.docx

Рядом появятся договор.masked.docx и договор.docx.vault.json (права 0600). Команда сама прогоняет контроль утечки и завершается с ненулевым кодом, если что-то осталось.

3. Отправить в модель только замаскированный текст. Либо полным кругом:

python3 -m pii_shield.cli ask договор.docx -p "Найди риски по срокам оплаты"

Эта команда сама маскирует, проверяет, отправляет и восстанавливает ответ. Провайдер задаётся флагом --provider anthropic|openai|deepseek|ollama.

4. Восстановить. Для ответа модели подстановка уже сделана. Замаскированный файл возвращается к исходному виду так:

python3 -m pii_shield.cli unmask договор.masked.docx --vault договор.docx.vault.json

Если работаешь из кода, а не из терминала

Когда нужно вплести обезличивание в свой пайплайн, важно одно: между текстом и сетью обязан стоять assert_clean.

from pii_shield import mask_text, assert_clean, unmask_text

r = mask_text(raw_text)          # r.text — с тегами, r.vault — сейф
assert_clean(r.text)             # LeakGuardError, если что-то осталось
answer = call_model(r.text)      # наружу уходит только маска
final, unknown = unmask_text(answer, r.vault, mode="canonical")

Read the full file on GitHub · 159 lines

Files

What ships with it

3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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.

1mo ago First seen · 159 lines · 296 tokens per session scan A 527999ef097c

Subscribe to this mod's changes

docs-masked is a skill published in the GitHub repository kpshinnik/docs_masked (21 stars, last pushed 2mo ago), licensed MIT. It adds 296 tokens to every session and 1,714 once invoked, about $0.0012 per session on Opus 5.5. 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 2026-08-30.

Documents LLM Security anonymization claude docx gdpr mcp

docx

Read, create, and convert Microsoft Word (.docx) documents — extract text and tables, build reports from markdown/JSON, and export to PDF.

zhixuli0406/DuDuClaw · 34 tokens

rediger-doc

À utiliser pour rédiger, réécrire, auditer ou mettre à jour de la documentation - un README, une documentation technique pour développeurs (architecture, guide, how-to, référence, runbook, ADR, guide de migration, changelog) ou une documentation métier/fonctionnelle pour utilisateurs non techniques (guide utilisateur…

Athroniaeth/piighost · 104 tokens

docs-masked