Skip to main content
Glama
openhoat

aik-mcp

by openhoat
OverviewSchemaRelated ServersScoreDiscussions
TypeScript
Local

aik-mcp turns a directory of Markdown files into a live, queryable knowledge base for any MCP-compatible AI agent — opencode, Claude Code, Cline, Codex, GitHub Copilot, and more.

Write your team's conventions, reusable workflows, agent prompts, and project templates as plain .md files with YAML frontmatter. aik-mcp serves them on demand — your agent can discover, read, search, and install them at runtime, across any project.

No database. No API to build. Just Markdown.

Features

Icon

Feature

Why it matters

📝

Knowledge as Markdown

Plain .md files with frontmatter. Version them with git. Review them in PRs.

⚡

Zero config

npx aik-mcp runs immediately. Point it at a folder of Markdown files. Done.

🔎

Full-text search

Fuzzy search across every rule, skill, and template — powered by Fuse.js.

📦

Install on demand

Push knowledge directly into your agent's runtime config with a single tool call.

👀

Live sync

A file watcher detects changes instantly. No restart. No downtime.

🔌

Universal MCP

Works with opencode, Claude Code, Cline, Codex, Copilot, and any MCP-compatible client.

Related MCP server: MinerU Document Explorer

Quick start

1. Create a rule

mkdir -p my-knowledge/rules
cat > my-knowledge/rules/typescript.md << 'EOF'
---
title: TypeScript Conventions
description: Coding standards for TypeScript projects
tags: [typescript, conventions]
version: "1.0.0"
compatibility: [opencode, claude-code, cline, codex, copilot]
---

## TypeScript Conventions

- Use explicit types for public API surfaces
- Prefer `interface` over `type` for object shapes
- Use `const` assertions for literal values
EOF

2. Start the server

AIK_CONTENT_DIR=./my-knowledge npx aik-mcp

3. Ask your agent

"Find and apply the TypeScript conventions rule for this project."

Your agent calls aik_search, reads the rule, and applies it — all transparently through MCP.

Docs

Full documentation is available at openhoat.github.io/aik-mcp.

How it works

graph LR
    Agent[AI Agent<br>opencode / Claude Code / Cline / Codex / Copilot] -->|MCP JSON-RPC| Server(aik-mcp)
    Server --> Store[Content Store<br>in memory]
    Store --> Files[Markdown files<br>rules/ skills/ workflows/ ...]
    Server --> Watcher[File Watcher<br>live sync on change]
    Server --> Tools[MCP Tools<br>list, get, search, write,<br>install, uninstall]

Your agent speaks MCP on one side. aik-mcp speaks your file system on the other. Everything is cached in memory for fast lookups, and a file watcher keeps the cache up to date.

Client configuration

opencode

Add to opencode.jsonc or .opencode/opencode.jsonc in your project:

{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "aik": {
      "type": "local",
      "command": ["npx", "-y", "aik-mcp"],
      "enabled": true,
      "environment": {
        "AIK_CONTENT_DIR": "/path/to/your/knowledge",
        "LOG_LEVEL": "info"
      }
    }
  }
}

Claude Code

Add to .mcp.json or ~/.claude/settings.json:

{
  "mcpServers": {
    "aik": {
      "command": "npx",
      "args": ["aik-mcp"],
      "env": {
        "AIK_CONTENT_DIR": "/path/to/your/knowledge",
        "LOG_LEVEL": "info"
      }
    }
  }
}

Cline

Add to cline.json or project .mcp.json:

{
  "mcpServers": {
    "aik": {
      "command": "npx",
      "args": ["aik-mcp"],
      "env": {
        "AIK_CONTENT_DIR": "/path/to/your/knowledge",
        "LOG_LEVEL": "info"
      }
    }
  }
}

Codex

Add to ~/.codex/config.toml or project .codex/config.toml:

[mcp]
"aik" = { command = ["npx", "aik-mcp"], env = { AIK_CONTENT_DIR = "/path/to/your/knowledge", LOG_LEVEL = "info" } }

GitHub Copilot

Configure the MCP server in your IDE settings (VS Code, JetBrains, etc.) and add project instructions to .github/copilot-instructions.md:

## MCP servers

aik-mcp provides knowledge management. Use `aik_list`, `aik_get`, `aik_search`, `aik_install`, and related tools to manage rules, skills, workflows, and templates.

Tip: Set AIK_CONTENT_DIR to a shared path (Dropbox, git repo, team NAS, etc.) to use the same knowledge base across projects and agents.

Content structure

Content items are organized by category:

Directory

Purpose

rules/

Coding standards, conventions, quality gates

skills/

Reusable instruction blocks (prompts, recipes)

workflows/

Multi-step process definitions

agents/

Specialized agent configurations

commands/

Custom CLI command definitions

templates/

File and project scaffolding

Each file is a Markdown document with YAML frontmatter:

---
title: "My Rule"
description: "What this rule enforces"
tags: [tag1, tag2]
version: "1.0.0"
compatibility: [opencode, claude-code, cline, codex, copilot]
---

## My Rule

Content here...

MCP tools

Tool

Description

aik_list

List content items, optionally filtered by category or tag

aik_get

Retrieve a specific item by path (e.g. rules/typescript)

aik_search

Full-text fuzzy search across all content

aik_write

Create or update a content item from the agent

aik_delete

Delete a content item

aik_install

Install an item into the project's agent config

aik_reinstall

Reinstall the latest version of an installed item

aik_uninstall

Remove an installed item from the project

aik_uninstall_all

Remove all aik-installed items from the project

aik_list_installed

List items currently installed in the project

Resources

URI

Description

aik://{category}

List all items in a category (e.g. aik://rules)

aik://search?q=...

Search items by keyword

CLI options

Flag

Default

Description

--http

—

Start in HTTP/SSE mode instead of stdio

--port <n>

3456

HTTP server port (only with --http)

--no-watch

—

Disable file watching

Environment variables

Variable

Default

Description

AIK_CONTENT_DIR

.

Path to the content directory

LOG_LEVEL

info

Log level: trace, debug, info, warn, error, silent

Development

npm install
npm run build
npm run test
npm run qa

Scripts

Script

Description

npm run build

Compile TypeScript to build/

npm test

Run Vitest test suite

npm run qa

Lint + format check (Biome + markdownlint)

npm run qa:fix

Auto-fix lint and formatting issues

npm run typecheck

TypeScript type checking (tsc --noEmit)

npm run validate

Full pipeline: qa → typecheck → build → test

Contributing

Contributions are welcome! Open an issue or submit a PR.

See the changelog for release history.

Full documentation at openhoat.github.io/aik-mcp.

License

MIT

Install Server
A
license - permissive license
A
quality
A
maintenance

How are these scores calculated?

Maintenance

–Maintainers
–Response time
1wRelease cycle
5Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Tools

View all tools

Related MCP Servers

View all related MCP servers

Related MCP Connectors

  • Darwin RAG

    Local-first RAG engine with MCP server for AI agent integration.

  • agentdocs-mcp

    MCP server for AgentDocs (agentdocs.eu): read, search, write, comment on & share Markdown docs.

  • mcp

    Your memory, everywhere AI goes. Build knowledge once, access it via MCP anywhere.

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/openhoat/aik-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server