Skip to content This library is free, offered as dana. If it helps your work, donate here →
penguin tree ai
← back to libraryDonate here

This is one of 3,622 free prompts in the penguin tree ai library. Sign up once and the whole library opens, with new prompts by email.

open the library →

Plain language translator for legal & compliance requirements

Translates dense legal, regulatory, or compliance language into plain, direct, inclusive product copy that preserves legal accuracy while being readable at a 6th-8th grade level. Delivers translated copy with a…

download .md

Purpose

Convert complex legal or compliance requirements into plain, direct, inclusive UI copy that a regular user can understand — without stripping the legal meaning that makes the copy compliant. Solves the tension product teams face between "legal signed off on this wall of text" and "users can't read it, don't consent meaningfully, and abandon the flow."

When to Use

Invoke this skill when the user asks to:

  • Rewrite terms of service, privacy policy excerpts, or EULA language for a product screen
  • Turn a consent, opt-in, or permission request into readable UI copy
  • Simplify a required disclosure (financial, health, data, GDPR/CCPA) for an interface
  • Make a compliance-driven error message, warning, or notice understandable
  • Translate a regulator's required language into onboarding, checkout, or settings copy
  • Reduce reading level or legalese in copy that legal or compliance owns

Inputs Required

Before starting, collect:

  • Legal source text: The exact legal, regulatory, or compliance language to translate (paste it in full)
  • UI context: Where the copy will appear (consent modal, checkout screen, onboarding step, error state, settings page, tooltip)
  • Regulatory regime: Any governing framework — GDPR, CCPA, HIPAA, TILA, FTC, SEC, or "unspecified"
  • Target reading level: Grade level goal (default: 7th-8th grade / Flesch-Kincaid 7-8)
  • Character or space constraints: Any UI limits (max characters, single sentence, button label length)
  • Terms that must stay verbatim: Any legally required wording legal has flagged as non-negotiable
  • Brand voice: Tone reference — neutral, warm, formal, or a linked voice guide

If regulatory regime or verbatim terms are unspecified, translate conservatively and flag every phrase where meaning could shift for legal review.

Instructions

Step 1: Parse the Legal Source Into Discrete Obligations

Break the source text into its atomic parts before rewriting anything:

  • Rights: What the user is entitled to
  • Obligations: What the user must do or agree to
  • Permissions requested: What the company is asking to do
  • Consequences: What happens if the user does or doesn't act
  • Disclosures: Facts that must be stated (fees, data use, risks)

List each part in one clause. This is your meaning inventory — nothing in the translation can add to or subtract from it.

Step 2: Identify Load-Bearing Legal Terms

Scan for words and phrases where the specific term carries legal weight and cannot be paraphrased:

  • Defined terms ("Personal Data," "Processing," "Material Adverse Effect")
  • Terms of art with regulatory meaning ("consent," "opt-in," "arbitration," "waive")
  • Quantities and timeframes that are legally binding (percentages, deadlines, dollar amounts)
  • Named parties and jurisdictions

Mark each as KEEP VERBATIM or SAFE TO SIMPLIFY. When uncertain, mark KEEP VERBATIM and add it to the flag list.

Step 3: Translate Using Plain Language Techniques

Rewrite each obligation applying these techniques in order:

  • Active voice + direct address: "We collect your location" not "Location data may be collected"
  • Second person: Speak to "you," not "the user" or "the data subject"
  • Common words for legalese: use/utilize, buy/purchase, before/prior to, end/terminate, if/in the event that
  • Short sentences: one idea per sentence, target 15-20 words max
  • Front-load the point: lead with what matters to the user, put conditions after
  • Concrete over abstract: "We keep your data for 90 days" not "retention periods apply"
  • Positive framing: state what will happen, not just what won't

Step 4: Apply Inclusive Language Standards

Check the translated copy for:

  • No idioms or figures of speech that don't translate across cultures
  • No assumptions about ability, family structure, gender, or literacy
  • Neutral, non-alarming tone for warnings (informative, not shaming)
  • Accessibility: screen-reader-friendly phrasing, no meaning conveyed by punctuation alone

Step 5: Verify Meaning Preservation

Compare the translation line-by-line against the Step 1 meaning inventory:

  • Every right, obligation, permission, consequence, and disclosure is still present
  • No new promise or limitation was accidentally introduced
  • No verbatim term from Step 2 was paraphrased away
  • Note any place where plain language forced a genuine trade-off, so legal can review

Step 6: Fit to UI Constraints and Score

  • Trim to the character/space limit while keeping every load-bearing element
  • Run a reading-level estimate (Flesch-Kincaid grade) and report it
  • If the copy is still above the target level, produce one tighter variant

Output Format

  • Structure: Original text shown, then the plain-language translation, then supporting analysis
  • Length: 400-800 words depending on source complexity
  • Format: Structured document with labeled sections; a two-column before/after table for the core translation
  • Sections: 1. Meaning Inventory — the discrete obligations extracted from the source 2. Plain-Language Translation — the final UI-ready copy (and a tighter variant if needed) 3. Before / After Table — original clause beside its translation 4. Verbatim Flags — terms kept unchanged and why 5. Legal-Review Flags — any place meaning could be read differently, for sign-off 6. Reading Level — estimated grade level of the translation
  • Tone: Clear, direct, warm-neutral in the copy itself; precise and cautious in the analysis

Quality Checklist

Before delivering output, verify:

  • [ ] Every item in the Meaning Inventory appears in the final translation
  • [ ] All load-bearing legal terms are either kept verbatim or explicitly flagged
  • [ ] The translation uses active voice and second person throughout
  • [ ] Reading level is reported with a specific grade number and meets or is flagged against the target
  • [ ] The copy fits the stated character/space constraint
  • [ ] Every plain-language trade-off is listed under Legal-Review Flags rather than hidden

Notes

  • This skill produces copy for review, not final legal sign-off — always route the Legal-Review Flags to counsel before shipping.
  • The most common failure is over-simplifying a defined term into a synonym that changes scope; when in doubt, keep the term and add a plain-language gloss next to it.
  • For consent flows specifically, meaningful consent requires the user to understand what they're agreeing to — a translation that's readable but vague can be less compliant than the original, not more.
  • When character limits are brutal (button labels, tooltips), lead with the single most legally material fact and link out to the full text rather than compressing meaning into ambiguity.

How to use this skill

This file is a Claude Code skill in Anthropic's standard format — markdown with YAML frontmatter at the top.

Installing in Claude Code:

  • ~/.claude/skills/<folder-name>/SKILL.md — auto-loads in every session anywhere on your machine (recommended for skills you'll use across multiple projects)
  • <your-project>/.claude/skills/<folder-name>/SKILL.md — auto-loads only when working in that project (recommended for project-specific skills)

Restart Claude Code or open a new session to register.

Using in Claude.ai, ChatGPT, Gemini, or other AI tools:

  1. Drag this file into a new chat, or copy and paste its contents
  2. Send your task or request as your next message — the AI will apply the skill where it fits

For Claude Code skill details: https://docs.anthropic.com/en/docs/claude-code/skills