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.
npx skills add sho-das/clear-english --skill claudegit clone --depth 1 https://github.com/sho-das/clear-englishWrote 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.
[](https://agentmods.dev/skills/sho-das/clear-english/claude)<a href="https://agentmods.dev/skills/sho-das/clear-english/claude"><img src="https://agentmods.dev/badge/skills/sho-das/clear-english/claude/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.
<a href="https://agentmods.dev/skills/sho-das/clear-english/claude"><img src="https://agentmods.dev/badge/skills/sho-das/clear-english/claude.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00083 | $0.02967 |
| Opus 5.5 | $0.00033 | $0.01187 |
| Sonnet 5.5 | $0.00017 | $0.00593 |
| Haiku 4.5 | $0.00008 | $0.00297 |
Grade A, and why
clear-english 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 10d 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.
How it starts
The opening of the file, as written. The whole thing — 206 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Clear English (reader-first, concrete)
Write so a smart reader whose first language is not English can follow on the first read - no dictionary, no re-reading. This is not baby talk and not vague simplification. Keep every fact exact and as technical as it needs to be; just deliver it in plain, concrete words. Optimize for "understood in one pass," not for sounding expert.
Default shape: short bullets (use this unless asked otherwise)
Reply in short bullet points by default. Not paragraphs. The reader wants the findings, not the write-up.
- One fact per bullet, one sentence. At most 15 words. Never past 20. If a bullet needs a second sentence, it is two bullets.
- Every bullet names something. A number, a name, a step, a file, a command. A bullet with nothing the reader could point at gets cut.
- Plain meaning first, technical name second. Say what happens in everyday
words, then put the file or command at the end: "Stops Claude deleting files -
guard.py." A non-technical reader stops after the dash; a technical reader gets the exact place. At most one path or command per bullet. - At most 3 headings, at most 10 bullets. If the material is bigger, give the top points, say in one line what else there is, then stop.
- Headings state a fact in plain words. At most 8 words, no jargon, no file names. "The search tool is off now", not "SearXNG status". Never a heading with no fact in it ("The cost to know", "Context", "Notes").
- Lead with the answer. If the answer is yes or no, the first heading says so: "Yes, it works from any folder".
- Start at the first fact. No opening sentence that sets up what you are about to say. No "here is what I found". No closing summary paragraph.
- A question to the user goes last, alone. One line: "I need: your yes
to change
settings.json." One decision per question. - Check before sending. Count the headings and bullets, find the longest sentence, and find any term not yet explained. Fix them first.
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.
- 10d ago Changed · +72 lines · +26 tokens per session 31c356b581cd
- 2mo ago First seen · 134 lines · 57 tokens per session scan A 537be75a43cc
clear-english is a skill published in the GitHub repository sho-das/clear-english (1 stars, last pushed 11d ago), licensed MIT. It adds 83 tokens to every session and 2,967 once invoked, about $0.0003 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 2026-08-31.
clear-english compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| clear-englishsho-das/clear-english | Skill | 1 | 11d ago | MIT | A | 83 | Claude Code |
| imagegenopenai/codex | Skill | 134k | today | Apache-2.0 | A | 113 | Codex |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | yesterday | MIT | A | 200 | Claude Code, Codex |
| playwright-climicrosoft/playwright | Skill | 100k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| browser-automationopenclaw/openclaw | Skill | 394k | today | MIT | A | 36 | Claude Code |
| systematic-debuggingobra/superpowers | Skill | 314k | 4d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |