- Home
- Skills
- Productivity
- Nativish
Works with the AI Agents you already use
Nativish
Inline English coaching for Claude Code: before each reply, see your message corrected and every fix explained.
Free
Nativish
Example session with this skill installed
the tests is failing after i update the config, can you explane why
- Read your context and instructions
- Compiled the nativish
✏️ The tests are failing after I updated the config. Can you explain why?
the tests is→ the tests are · subject-verb agreement
update→ updated · verb tense
i→ I · lowercase pronoun
explane→ explain · spelling
Which config did you change? If it touched the test environment, that's the usual suspect. Paste the failing output and I'll trace it.
A clean message gets one line instead:
💪 Sharp grammar — you're leveling up.
Connects securely to your tools. The creator never sees your data.
About this skill
Before Claude answers, nativish rewrites your message in correct English, then lists each fix: the mistake struck through, the correction in bold, and the reason in a few words. Then Claude answers as usual.
What it does
- Corrects real mistakes: grammar, tense, articles, wrong words, casing like
github→GitHub. - Ignores chat style: lowercase starts, missing periods,
dont,pls. - Skips non-prose: slash commands, short replies like
ok, paths, versions, mostly non-Latin text. - Praises clean messages in one line.
Send nativish off / nativish on to toggle; it resets on /clear.
Get the full version
Standalone, the skill coaches only when Claude loads it. To get coaching on every message, install the free plugin:
/plugin marketplace add denys-petryniak/nativish
/plugin install nativish@nativish
Source: https://github.com/denys-petryniak/nativish
No network calls, no telemetry, no dependencies.
Built for Claude Code; other agents are untested. As a standalone skill, it coaches only when the agent loads it, not on every message; the Claude Code plugin fixes that. Messages that are mostly a task (e.g. "clean up /tmp") may skip coaching on stronger models. Pasted code and logs get coached too, so expect the odd false flag. On long messages, the rewrite stops after a few sentences, but fixes for the rest still appear.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 14 days ago
- Free with an Agensi account
Permissions
No special permissions declared or detected