Agent Skills English Productivity Pack
Five instruction-only skills for proofreading, notes, and research
About
A compact MIT-licensed pack of five agent skills for everyday knowledge work: tech-writing-proofread (checklist proofreading that flags defects without rewriting your voice), meeting-notes (raw notes into structured minutes with owners and deadlines — refuses to invent dates or names), git-commit-message (conventional commits from staged changes), code-review-checklist (five-axis review capped at 10 comments), and deep-research-framework (tiered sources, cross-verification, explicit uncertainty). Every skill is a single instruction-only SKILL.md — no scripts, no binaries, 283 lines total, auditable in one sitting. Install with npx skills add alapha888/agent-skills-en. Thin-vetting caveat: 1 star and 27 commits since its October 2026 Show HN launch — real, maintained content, but essentially no community track record yet.
What it can do
- Checklist proofreading for technical writing
- Meeting notes to structured minutes
- Conventional commit message generation
- Five-axis code review checklist
- Deep research framework with source tiering
- Instruction-only: no scripts, no binaries, MIT licensed
Setup
Run npx skills add alapha888/agent-skills-en for all five skills, or install one via SkillMD (e.g. npx skillmds@latest add alapha888/meeting-notes). Tell your agent what you want done and point at the SKILL.md if it doesn't pick it up.
Prerequisites
Node.js for the Skills CLI, or any agent that reads SKILL.md.
Use cases
Works with
Trust score — 6/10
Scored 6/10 (metrics tier, metrics-v1). Evidence: Live GitHub page read this session: 1 star, MIT license, 27 commits; install commands copied verbatim from README.; Five real instruction-only skills (283 lines total) with worked examples verified in README; meeting-notes' no-fabrication rule is a genuine design strength.; Thin vetting: 1 star since October 2026 Show HN launch — metrics tier with explicit track-record caveat..
★ 1 GitHub stars
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.