The problem
Style guides are long. Deadlines are short.
Readability tools flag problems but do not fix them.
Rewrites change meaning in legal and eligibility text.
06
Plain English, enforced. Every edit explained.
UCD Writer rewrites interface copy to GOV.UK style and your chosen reading level, then shows you each change, the rule behind it, and why.
Or get all nine agents in the UCD Team bundle.
Style guides are long. Deadlines are short.
Readability tools flag problems but do not fix them.
Rewrites change meaning in legal and eligibility text.
Scores every sentence for reading level
Flags long sentences, complex words, passive voice, jargon and idioms
Rewrites to grade 8, grade 9 or reading age 9
Protects legal, policy and money text from rewording
Explains every edit in one plain sentence
Get this on Gumroad and download the zip.
Unzip it into a project folder and open it in your AI tool.
Say `check this against GOV.UK style: <file or text>`.
It works one stage at a time and stops for your yes.
Reading-level check, Style and inclusion check
Rewrite to standard, Explain the edit
Final copy
GOV.UK style guide and a grade 8, grade 9 or reading age 9 ceiling.
Every claim cites its source. Nothing is written to your tools without your approval.
Product owners, delivery managers and writers on UK public sector services.
An AI tool that reads project instructions: Claude Code, Cursor, or a tool that reads AGENTS.md. Claude Projects and custom bots work too, with copy-and-paste outputs.
Your own AI subscription or API key. Nothing to host.
Optional connections: Reading-age script.
| File | What it does |
|---|---|
00-root-instructions.md | Who the agent is, the rules, the gates |
01-analyse.md | Stage 1: Analyse |
02-rewrite.md | Stage 2: Rewrite |
03-finalise.md | Stage 3: Finalise |
04-tools.md | Connections and permissions |
05-manifest.md | Which skill uses which tool |
06-adapters.md | How to load it in your tool |
LICENCE.md | What you may and may not do |
README.md | Start here: set-up in two minutes |
guardrails.md | Risks, controls and sign-off |
memory.md | What is kept and where |
adapters/AGENTS.md | Adapter for your AI tool |
adapters/CLAUDE.md | Adapter for your AI tool |
adapters/ucd-writer.mdc | Adapter for your AI tool |
adapters/chat-system-prompt.md | Adapter for your AI tool |
evals/eval-suite.md | Tests and pass marks |
rules/gov-uk-style.md | Standard the agent checks against |
rules/inclusive-language.md | Standard the agent checks against |
rules/words-to-avoid.md | Standard the agent checks against |
samples/01-poor-eligibility.md | Synthetic sample input |
samples/02-error-messages.md | Synthetic sample input |
samples/03-start-page.md | Synthetic sample input |
tools/readability.py | Local helper script |
Designed before it was built
This agent was designed with the five-question method: purpose, boundaries, what it touches, when a person steps in, and how you know it works. [Read the design](./anatomy).
Designed by Vijay Kandhalu.