Summary
This is Matt Pocock's communication framework turned into agent instructions. It fights the natural tendency of LLMs to be verbose by enforcing brutal concision: fragments over sentences, tables over paragraphs, action steps always last. The before/after examples are immediately useful (replacing "The user will be able to" with "User can" adds up fast). Works especially well for planning outputs where you're scanning terminal history bottom-up. The SpecKit integration markers like [C§3.1] and [US1] show this came from a real workflow. If your agents write novels when you need bullet points, this fixes that.
Install to Claude Code
npx -y skills add tzachbon/smart-ralph --skill communication-style --agent claude-codeInstalls into .claude/skills of the current project.
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
An agent that runs the SEO playbooks that move rankings and ships PRs you control.
Get founding access →
Connect Claude to +800M contacts, +150M companies. Find & Enrich leads in chat.
Try For Free →Files
SKILL.md
Select a file.
Featured
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →An agent that runs the SEO playbooks that move rankings and ships PRs you control.
Get founding access →Connect Claude to +800M contacts, +150M companies. Find & Enrich leads in chat.
Try For Free →
