Skip to content

Commit 1ee5258

Browse files
rohitg00rohitg00
authored
docs: rewrite with clean, concise style (#43)
* docs: rewrite documentation with clean, concise style Rewrote all core docs pages (index, installation, quickstart, rest-api, mcp-server, python-client) with problem/solution framing, progressive code examples, and conversational tone. Updated OpenAPI descriptions and api.html meta tags to match. 366 additions vs 722 deletions. * docs(mcp): remove stale skill count, add list_categories tool name Replace hard-coded "15,000+" with non-specific wording and explicitly name the list_categories tool in the Browse categories section.
1 parent db38c90 commit 1ee5258

File tree

8 files changed

+366
-722
lines changed

8 files changed

+366
-722
lines changed

‎docs/fumadocs/content/docs/index.mdx‎

Lines changed: 95 additions & 115 deletions
Original file line numberDiff line numberDiff line change
@@ -1,116 +1,76 @@
11
---
2-
title: SkillKit Documentation
3-
description: Universal CLI for managing AI agent skills across 32 coding platforms
2+
title: SkillKit
3+
description: Universal skill management for AI coding agents
44
---
55

66
# SkillKit
77

8-
SkillKit is a universal command-line interface (CLI) and programmatic toolkit for managing AI agent skills across **32 AI coding platforms**.
8+
Skills make AI coding agents smarter. But every agent uses a different format — Claude Code wants `.claude/skills/`, Cursor uses `.mdc`, Copilot expects `.github/skills/`. You end up rewriting the same skill three times, or locking into one platform.
99

10-
**Current Version:** v1.12.0
10+
SkillKit fixes this. Write a skill once, deploy it to all 32 agents.
1111

12-
## The Problem
13-
14-
Each AI coding agent uses different skill formats and directory structures:
15-
- Claude Code uses `.claude/skills/` with SKILL.md
16-
- Cursor uses `.cursor/skills/` with .mdc format
17-
- GitHub Copilot uses `.github/skills/`
18-
19-
Without SkillKit, you either rewrite skills for each agent, lock into one platform, or abandon skills entirely.
20-
21-
## The Solution
22-
23-
SkillKit provides a **"write once, deploy everywhere"** platform. Maintain a single skill definition that automatically translates across all 32 platforms.
24-
25-
## Quick Install
12+
## Get Started
2613

2714
```bash
28-
# Run instantly with npx (no install required)
2915
npx skillkit@latest
16+
```
17+
18+
That's it. No global install needed. Or if you prefer:
3019

31-
# Or install globally for frequent use
20+
```bash
3221
npm install -g skillkit
3322
skillkit init
3423
skillkit install anthropics/skills
3524
skillkit sync
3625
```
3726

38-
## Core Features
39-
40-
| Feature | Description |
41-
|---------|-------------|
42-
| **Cross-Platform Translation** | Automatic conversion between 32 agent formats |
43-
| **Smart Recommendations** | AI-powered skill suggestions based on your tech stack |
44-
| **Session Memory** | Persistent learning that survives across sessions |
45-
| **Primer** | Auto-generate agent instructions from your codebase |
46-
| **Mesh Network** | Multi-machine agent distribution with encrypted P2P |
47-
| **Inter-Agent Messaging** | Send messages between AI agents across your network |
48-
| **Team Collaboration** | Git-based `.skills` manifest for sharing |
49-
| **Workflow Orchestration** | Compose skills into multi-step automated workflows |
50-
| **Skill Testing** | Built-in test framework with assertions |
51-
| **CI/CD Integration** | GitHub Actions, GitLab CI, and pre-commit hooks |
52-
| **REST API Server** | Runtime skill discovery via HTTP on port 3737 |
53-
| **MCP Server** | Agent-native discovery for Claude Desktop, Cursor |
54-
| **Python Client** | Async Python SDK for the REST API |
55-
| **TypeScript API** | Full programmatic access to all features |
56-
| **Interactive TUI** | Beautiful terminal interface with real-time updates |
57-
58-
## Advanced Capabilities
27+
Four commands. Your agents now have skills for PDF processing, code review, and more.
5928

60-
### Session Memory System
29+
## What Can You Do?
6130

62-
Your AI agents learn patterns during sessions—then forget everything. SkillKit captures these learnings and makes them permanent.
31+
### Install skills from anywhere
6332

6433
```bash
65-
skillkit memory compress # Compress learnings
66-
skillkit memory search auth # Search knowledge
67-
skillkit memory export auth-patterns # Export as skill
34+
skillkit install anthropics/skills # GitHub
35+
skillkit install gitlab:team/skills # GitLab
36+
skillkit install ./my-local-skills # Local
6837
```
6938

70-
[Learn more →](/docs/memory)
39+
### Translate between agents
7140

72-
### Primer - Auto-Generate Instructions
73-
74-
Automatically generate CLAUDE.md, .cursorrules, and agent instructions by analyzing your codebase.
41+
Write for Claude, deploy to Cursor:
7542

7643
```bash
77-
skillkit primer # Analyze project
78-
skillkit primer --all-agents # Generate for all 32 agents
44+
skillkit translate my-skill --to cursor
45+
skillkit translate --all --to windsurf # All skills at once
7946
```
8047

81-
[Learn more →](/docs/primer)
82-
83-
### Mesh Network
48+
### Get smart recommendations
8449

85-
Run agents across multiple machines with encrypted P2P communication.
50+
SkillKit reads your `package.json`, detects your stack, and suggests relevant skills:
8651

8752
```bash
88-
skillkit mesh init # Initialize mesh node
89-
skillkit mesh discover # Find peers on LAN
90-
skillkit mesh security init # Setup Ed25519 encryption
53+
skillkit recommend
54+
# 92% vercel-react-best-practices
55+
# 87% tailwind-v4-patterns
56+
# 85% nextjs-app-router
9157
```
9258

93-
[Learn more →](/docs/mesh)
59+
### Discover skills at runtime
9460

95-
### Inter-Agent Messaging
96-
97-
Send messages between AI agents across your mesh network.
61+
Start an API server and let agents find skills on demand:
9862

9963
```bash
100-
skillkit message send # Send to another agent
101-
skillkit message inbox # Check your inbox
64+
skillkit serve
65+
# Server running at http://localhost:3737
10266
```
10367

104-
[Learn more →](/docs/messaging)
105-
106-
### Runtime Skill Discovery
107-
108-
Start a REST API or MCP server for runtime skill discovery by agents and applications.
109-
11068
```bash
111-
skillkit serve # REST API on :3737
69+
curl "http://localhost:3737/search?q=react+performance"
11270
```
11371

72+
Or use MCP for native agent integration:
73+
11474
```json
11575
{
11676
"mcpServers": {
@@ -119,55 +79,75 @@ skillkit serve # REST API on :3737
11979
}
12080
```
12181

122-
[REST API →](/docs/rest-api) · [MCP Server →](/docs/mcp-server) · [Python Client →](/docs/python-client)
82+
[REST API](/docs/rest-api) | [MCP Server](/docs/mcp-server) | [Python Client](/docs/python-client)
12383

124-
## Architecture
84+
## How It Works
12585

126-
| Tier | Components |
127-
|------|------------|
128-
| **User Interfaces** | CLI (Clipanion v4), TUI (OpenTUI/Solid.js), Documentation |
129-
| **APIs** | @skillkit/api (REST/Hono), @skillkit/mcp (MCP/stdio) |
130-
| **Business Logic** | @skillkit/core, @skillkit/agents (32 adapters) |
131-
| **Data Layer** | @skillkit/memory (CozoDB), @skillkit/mesh (WebSocket/UDP) |
132-
| **External Systems** | GitHub/GitLab, Marketplace (15,000+ skills) |
86+
SkillKit sits between your skills and your agents. It handles format translation, discovery, and synchronization automatically.
13387

134-
## Supported Agents (32)
88+
| Layer | What It Does |
89+
|-------|-------------|
90+
| **CLI + TUI** | Install, translate, recommend, sync |
91+
| **APIs** | REST server on :3737, MCP via stdio |
92+
| **Core** | Translation engine, relevance ranker, skill parser |
93+
| **Data** | 15,000+ skills from 31 curated sources |
13594

136-
Claude Code, Cursor, Codex, Gemini CLI, OpenCode, GitHub Copilot, Windsurf, Antigravity, Amp, Clawdbot, Droid, Goose, Kilo Code, Kiro CLI, Roo Code, Trae, Cline, CodeBuddy, Continue, Crush, Factory, MCPJam, Mux, Neovate, OpenHands, Pi, Qoder, Qwen, Vercel, Zencoder, CommandCode, Universal
95+
## Supported Agents
13796

138-
[Full agent details →](/docs/agents)
97+
32 agents and counting:
13998

140-
## Popular Skill Sources
99+
**Claude Code** | **Cursor** | **Codex** | **Gemini CLI** | **OpenCode** | **GitHub Copilot** | **Windsurf** | Amp | Antigravity | Clawdbot | Cline | CodeBuddy | CommandCode | Continue | Crush | Droid | Factory | Goose | Kilo Code | Kiro CLI | MCPJam | Mux | Neovate | OpenHands | Pi | Qoder | Qwen | Roo Code | Trae | Vercel | Zencoder | Universal
141100

142-
| Repository | Skills |
143-
|------------|--------|
144-
| anthropics/skills | PDF, XLSX, DOCX processing |
145-
| vercel-labs/agent-skills | React/Next.js best practices |
146-
| nextlevelbuilder/ui-ux-pro-max-skill | UI/UX design patterns |
147-
| superdesigndev/superdesign-skill | Modern design systems |
101+
[Full agent details](/docs/agents)
148102

149-
## Get Started
103+
## Beyond Basic Skills
104+
105+
### Session Memory
106+
107+
Your AI agents learn patterns during sessions, then forget everything. SkillKit captures those learnings and makes them permanent.
108+
109+
```bash
110+
skillkit memory compress
111+
skillkit memory search "auth patterns"
112+
skillkit memory export auth-patterns
113+
```
114+
115+
### Primer
116+
117+
Auto-generate CLAUDE.md, .cursorrules, and agent instructions by analyzing your codebase:
118+
119+
```bash
120+
skillkit primer --all-agents
121+
```
122+
123+
### Mesh Network
124+
125+
Distribute agents across machines with encrypted P2P:
126+
127+
```bash
128+
skillkit mesh init
129+
skillkit mesh discover
130+
```
131+
132+
### Team Collaboration
133+
134+
Share skills via a Git-committable `.skills` manifest:
135+
136+
```bash
137+
skillkit manifest init
138+
skillkit manifest add anthropics/skills
139+
git commit -m "add team skills"
140+
```
141+
142+
Team members run `skillkit manifest install` and they're in sync.
143+
144+
## Next Steps
145+
146+
- [Installation](/docs/installation) — Install and configure
147+
- [Quick Start](/docs/quickstart) — First project setup
148+
- [Commands](/docs/commands) — Full CLI reference
149+
- [REST API](/docs/rest-api) — Runtime discovery server
150+
- [MCP Server](/docs/mcp-server) — Agent-native discovery
151+
- [Python Client](/docs/python-client) — Python SDK
150152

151-
1. [Installation](/docs/installation) - Install and initialize
152-
2. [Quick Start](/docs/quickstart) - 5-minute guide
153-
3. [Commands](/docs/commands) - Full CLI reference
154-
4. [Skills](/docs/skills) - Skill format and lifecycle
155-
5. [API Reference](/docs/api-reference) - Programmatic access
156-
6. [REST API](/docs/rest-api) - Runtime discovery server
157-
7. [MCP Server](/docs/mcp-server) - Agent-native discovery
158-
8. [Python Client](/docs/python-client) - Python SDK
159-
160-
## Use Cases
161-
162-
- **Multi-Agent Developer** - Write skills once, use across Claude, Cursor, Windsurf, and more
163-
- **Team Lead** - Share skills via Git-based manifest, ensure team alignment
164-
- **New Project Starter** - Auto-generate agent instructions with Primer
165-
- **Enterprise Architect** - Distribute agents across multiple machines with Mesh Network
166-
167-
## Enterprise Features
168-
169-
- **Private Registries** - Host internal skills in private repos
170-
- **Git-Based Sync** - Skills live in your repo, push to share
171-
- **CI/CD Integration** - Auto-validate and deploy skills on every commit
172-
- **TypeScript API** - Full programmatic access for custom workflows
173-
- **Security** - Ed25519 encryption, XChaCha20-Poly1305, TLS 1.3
153+
**Current version:** v1.12.0 | [npm](https://www.npmjs.com/package/skillkit) | [GitHub](https://github.com/rohitg00/skillkit) | Apache-2.0

0 commit comments

Comments
 (0)