
Agentskb
- 9 repo stars
- Updated July 27, 2026
- Cranot/agentskb-mcp
agentskb is a MCP server that supplies AI agents with researched, canonical knowledge from official sources.
About
agentskb is an MCP server that surfaces pre-researched, source-aligned knowledge for coding agents. developers shipping Claude Code or Cursor workflows hit a wall when the model invents endpoints or mixes doc versions; agentskb is meant to supply canonical answers you can trust during implementation. Install the @agentskb/cli package, wire stdio transport in your MCP config, and the agent can query the knowledge base while you build integrations, agents, and automations. It is phase-specific to Build agent-tooling rather than a full research suite for market discovery. Pair it with your normal docs and skills when you need higher-confidence factual retrieval during long agent sessions.
- Researched knowledge corpus aimed at AI agent workflows
- Canonical answers grounded in official sources
- Published npm package @agentskb/cli with stdio MCP transport
- Repository agentskb-mcp on GitHub (Cranot)
- Reduces wrong SDK versions and deprecated API usage during codegen
Agentskb by the numbers
- Data as of Jul 28, 2026 (Skillselion catalog sync)
claude mcp add agentskb -- npx -y @agentskb/cliAdd your badge
Show developers this MCP server is listed on Skillselion. Paste this into your README.
| repo stars | ★ 9 |
|---|---|
| Package | @agentskb/cli |
| Transport | STDIO |
| Auth | None |
| Last updated | July 27, 2026 |
| Repository | Cranot/agentskb-mcp ↗ |
What it does
Give your agent a researched knowledge layer with canonical answers from official sources instead of hallucinated API details.
Who is it for?
Best when you're creating agent workflows and want a dedicated knowledge MCP alongside codegen tools.
Skip if: Market research, competitor analysis, or brand-new domains with no curated agentskb coverage.
What you get
MCP queries return researched canonical answers so implementations align with official documentation.
- MCP-accessible researched answers during build sessions
- Lower-risk API and SDK usage in generated code
- Repeatable knowledge lookups without manual doc tab switching
By the numbers
- Server version 0.1.1 with npm package @agentskb/cli at 0.1.0-beta.8
- Stdio transport type in server.json packages block
- Source repository https://github.com/Cranot/agentskb-mcp
README.md
AgentsKB MCP Server
Give Your AI Expert-Level Knowledge
Researched answers to any problem. Your AI, upgraded.
Installation
Add to your Claude Desktop or Cursor config:
{
"mcpServers": {
"agentskb": {
"command": "npx",
"args": ["@agentskb/cli", "mcp"]
}
}
}
No API key needed for free tier (300 queries/month).
What makes us different
Your AI searches. We research. That's the difference.
- Dedicated research per question - not a side task while debugging
- Official sources only - docs, specs, RFCs
- Would rather not answer than guess wrong
Tools
| Tool | Description |
|---|---|
ask_question |
Expert-quality answers with sources |
preflight_check |
See what verified knowledge exists for your task |
search_questions |
Browse Q&As by topic |
get_stats |
Coverage metrics |
Coverage
Growing knowledge base across 40+ developer domains:
- Databases: PostgreSQL, MySQL
- Frameworks: Next.js, React, Vue, FastAPI, Django, Laravel
- Infrastructure: Docker, Kubernetes, AWS
- And more being added continuously
Links
License
MIT
Recommended MCP Servers
How it compares
Curated agent knowledge MCP, not a live web browser or generic search plugin.
FAQ
Who is agentskb for?
Developers and small teams building with AI coding agents who need dependable factual lookups during integration work.
When should I use agentskb?
Use it during Build when the agent is wiring SDKs, APIs, or agent tools and you want official-source-aligned answers on demand.
How do I add agentskb to my agent?
Install @agentskb/cli from npm, configure stdio MCP transport in Claude Code or Cursor, and point your server list at the agentskb MCP entry.