agentleFS
miolamiomiolamio/agent-skills/skills/ru-textovod/SKILL.md

ru-textovod

Skill13 starsChanged 8 months agoMITView on GitHub

Checks Russian text for spelling, punctuation, and grammar via Textovod.com API. Use when user says "проверь текст", "проверь орфографию", "проверь пунктуацию", "автокоррекция", "исправь ошибки", "/ru-textovod", or after writing/editing Russian text. Three independent commands: autocorrection (default), spelling, punctuation. NOT for stylistic editing (use ru-editor).

  • Reads credentials
  • Sends data out

2 other projects use a version of this, 2 of them edited.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 2

    1. miolamio/claude-skills★ 28100% same
    2. eazieman88/ClaudeSkills★ 0100% same

    Wrote their own version · 0

    None found.

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository5 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 4 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Always use jq -n --arg to build JSON bodies — never interpolate text directly into JSON strings.

    line 24
  2. Distinctive:

    Always escape and validate text before sending to API.

    line 25
  3. Distinctive:

    Before any API call, check that the required env vars are set:

    line 39
  4. Very specific:

    Always: use jq -n --arg to safely embed text into JSON:

    line 118
  5. Very specific:

    Never do {"text": "$TEXT"} — this breaks on quotes, newlines, and special characters.

    line 124
  6. Distinctive:

    Before sending, verify text length:

    line 128
  7. Distinctive:

    If text exceeds the limit, inform the user and suggest splitting.

    line 136
  8. Very specific:

    Important: Do NOT send id in the request body — the API generates its own UUID.

    line 182

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Russian Text Proofreader (Проверка русского текста)

Important Rules

  • Default command is autocorrection (API v2 grammar). Use it unless the user explicitly asks for spelling or punctuation separately.
  • Never modify files without user permission. Show the corrected text and ask before applying changes.
  • This skill is for proofreading — spelling, punctuation, grammar. For stylistic editing (AI markers, informational style, voice), recommend ru-editor.
  • Always use jq -n --arg to build JSON bodies — never interpolate text directly into JSON strings.
  • Always escape and validate text before sending to API.

Prerequisites — API Key Setup

Three environment variables are required:

VariableAPIPurpose
TEXTOVOD_USER_IDv1User ID for spelling/punctuation
TEXTOVOD_API_KEYv1API key for spelling/punctuation
TEXTOVOD_API2_TOKENv2Bearer token for autocorrection

Check Credentials

Before any API call, check that the required env vars are set:

# For autocorrection (API v2)
echo "${TEXTOVOD_API2_TOKEN:+set}"

# For spelling/punctuation (API v1)
echo "${TEXTOVOD_USER_ID:+set} ${TEXTOVOD_API_KEY:+set}"

If Credentials Are Missing

Tell the user which variables are missing and how to set them:

Для работы нужны API-ключи Textovod.com.

Получите ключи на https://textovod.com/api и добавьте в ~/.zshrc:

export TEXTOVOD_USER_ID="your_user_id"
export TEXTOVOD_API_KEY="your_api_key"
export TEXTOVOD_API2_TOKEN="your_api2_token"

Затем перезапустите терминал или выполните: source ~/.zshrc

Alternatively, keys can be set in Claude Code settings (~/.claude/settings.json) under env.

Three Commands

1. Autocorrection (default)

Trigger: «проверь текст», «автокоррекция», «исправь ошибки», or no specific command.

Neural network-based. Fixes grammar, spelling, and punctuation in one pass. Best for general proofreading.

  • API: v2 grammar
  • Requires: TEXTOVOD_API2_TOKEN
  • Max: 35,000 characters
  • Reference: references/api-v2-grammar.md

2. Spelling

Trigger: «проверь орфографию», «spelling», «орфография».

Dictionary-based spelling check. Returns individual misspelled words with suggestions.

  • API: v1 spelling
  • Requires: TEXTOVOD_USER_ID + TEXTOVOD_API_KEY
  • Max: 100,000 characters
  • Reference: references/api-v1-spelling.md

3. Punctuation

Trigger: «проверь пунктуацию», «punctuation», «пунктуация», «запятые».

Rule-based punctuation check. Returns corrected text with punctuation fixes.

  • API: v1 punctuation
  • Requires: TEXTOVOD_USER_ID + TEXTOVOD_API_KEY
  • Max: 100,000 characters
  • Reference: references/api-v1-punctuation.md

Input Handling

Inline Text (primary)

User provides text directly in the message. Extract the text to check.

File Path

User provides a file path. Read the file content, then proceed as with inline text.

TEXT=$(cat "/path/to/file.txt")

JSON Escaping

Always use jq -n --arg to safely embed text into JSON:

JSON_BODY=$(jq -n --arg text "$TEXT" '{text: $text, lang: "ru"}')

Never do {"text": "$TEXT"} — this breaks on quotes, newlines, and special characters.

Character Limit Check

Before sending, verify text length:

CHAR_COUNT=$(echo -n "$TEXT" | wc -c)
# API v2: max 35000
# API v1: max 100000

If text exceeds the limit, inform the user and suggest splitting.

API Workflow

Each API has a different workflow:

Spelling (v1) — Immediate Response

Returns results in the submit response. No polling needed.

JSON_BODY=$(jq -n \
  --arg user_id "$TEXTOVOD_USER_ID" \
  --arg api_key "$TEXTOVOD_API_KEY" \
  --arg text "$TEXT" \
  '{user_id: $user_id, api_key: $api_key, text: $text, lang: "ru"}')

RESPONSE=$(curl -s -X POST https://textovod.com/api/spelling/user/add \
  -H "Content-Type: application/json" \
  -d "$JSON_BODY")

# result is a JSON string — parse with fromjson
echo "$RESPONSE" | jq -r '.result | fromjson | .[] | "\(.offset) \(.length) \(.mes) \(.rep[0])"'

Punctuation (v1) — Submit + Poll with text_id

# Submit
RESPONSE=$(curl -s -X POST https://textovod.com/api/punctuation/user/add \
  -H "Content-Type: application/json" -d "$JSON_BODY")
TEXT_ID=$(echo "$RESPONSE" | jq -r '.text_id')

# Poll (separate Bash calls, 3 sec apart, use text_id not id)
POLL_BODY=$(jq -n \
  --arg user_id "$TEXTOVOD_USER_ID" \
  --arg api_key "$TEXTOVOD_API_KEY" \
  --arg text_id "$TEXT_ID" \
  '{user_id: $user_id, api_key: $api_key, text_id: $text_id}')
RESULT=$(curl -s -X POST https://textovod.com/api/punctuation/user/status \
  -H "Content-Type: application/json" -d "$POLL_BODY")
# status 2 = ready

Autocorrection (v2) — Submit + Poll

Important: Do NOT send id in the request body — the API generates its own UUID.

# Submit (no id field!)
JSON_BODY=$(jq -n --arg text "$TEXT" '{text: $text, lang: "ru"}')
RESPONSE=$(curl -s -X POST https://textovod.com/api/v2/text/grammar \
  -H "Authorization: Bearer ${TEXTOVOD_API2_TOKEN}" \
  -H "Content-Type: application/json" -d "$JSON_BODY")
TASK_ID=$(echo "$RESPONSE" | jq -r '.id')

# Poll (separate Bash calls, 3 sec apart)
POLL_BODY=$(jq -n --arg id "$TASK_ID" '{id: $id}')
RESULT=$(curl -s -X POST https://textovod.com/api/v2/text/grammar/result \
  -H "Authorization: Bearer ${TEXTOVOD_API2_TOKEN}" \
  -H "Content-Type: application/json" -d "$POLL_BODY")
# Corrected text: jq -r '.messages[] | select(.role == "assistant") | .content'

Polling: Make each poll as a separate Bash tool call with sleep 3 at the start. Max 15 attempts.

Output Modes

See references/result-formatting.md for full formatting details.

Detailed Mode (default)

Show list of errors + corrected text:

### Найденные ошибки (N)

1. «неправильнoe» -> «правильное» (орфография)
2. «предложение без запятой» -> «предложение, с запятой» (пунктуация)

### Исправленный текст

[corrected text]

Quick Mode

Triggered by: «просто исправь», «без деталей», «только текст»

Исправлено ошибок: N

[corrected text]

Error Handling

ErrorAction
Missing API keysShow setup instructions (see Prerequisites)
Status 30 (insufficient balance)Tell user: «Недостаточно средств на балансе Textovod. Пополните на https://textovod.com»
Timeout (15 polls without result)Tell user: «Сервис не ответил за 45 секунд. Попробуйте позже или с текстом покороче.»
Rate limit (429)Wait 5 seconds, retry once
Text too longShow limit and suggest splitting
Empty textTell user: «Нет текста для проверки.»
Network errorShow error and suggest retrying

Integration with Other Skills

  • After checking — if text has stylistic issues (AI markers, bureaucratese, weak verbs), suggest: «Текст грамматически верен. Хотите стилистическую редактуру? Используйте ru-editor.»
  • After ru-editor — suggest: «Хотите проверить орфографию и пунктуацию? Используйте ru-textovod.»
  • After en-ru-translator-adv — suggest proofreading the translation with ru-textovod.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.