Skip to content
← Tools / Skills
Skill · .md packagenstung463/10 trustMetrics-checked
Kimi Agent Core Skills logo

Kimi Agent Core Skills

261-skill community packaging of the Kimi Agent core library

Visit official site →

About

A community packaging of the 261-skill core collection supplied with the Kimi Agent setup — writing, research, finance, product planning, engineering, data, documents (docx, pdf, xlsx, pptx), media, and productivity. Each skill is self-contained with a SKILL.md and supporting scripts, templates, or references, installable as a whole via the skills.sh CLI or one skill at a time by copying directories. The strength is sheer coverage: the widest skill set in this batch. The limitation is that it is not official — the repo warns individual skills carry different authorship and license terms, and some assume Kimi Agent conventions, so check licensing and dependencies before commercial use.

What it can do

  • 261 skills across writing, research, finance, engineering, data, documents, media, productivity
  • Document skills: docx, pdf, xlsx, pptx
  • Research skills: deep-research, academic-paper-reviewer
  • Marketing entry points: seo-audit, campaign-planner, brand-naming-lab
  • Skill discovery helpers: kimi-find-skills, kimi-skills-finder
  • Self-contained SKILL.md per skill with supporting files

Setup

npx skills add nstung463/kimi-agent-skills

Single skill: copy its directory from skills/ into your agent environment's skills folder, keeping SKILL.md with its supporting files

Prerequisites

None for the install; individual skills may document their own dependencies

Use cases

others

Works with

Claude CodeCodexCursorgemini-cliWindsurf

Trust score — 4/10

/10 trustMetrics-checked

Scored 4/10 (metrics tier, metrics-v1). Evidence: Community packaging, not official; 5 stars, 6 commits; mixed per-skill licensing requires checking before commercial use. License: No root license; repo warns individual skills carry their own licensing terms - unconfirmed.

★ 5 GitHub stars

How the trust score is built

Full methodology →
Adoption · stars/downloads, log-scaled vs peers
3.0/10
Maintenance · recency, releases, issue load
4.0/10
Openness · license + docs
4.0/10
Manual review · sanity read + red-flag check
6.0/10
  • · Community packaging, not official; 5 stars, 6 commits; mixed per-skill licensing requires checking before commercial use

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.