organisation.md
Turns any GitHub repository into your organisation's persistent context layer, enabling AI agents to read, search, and propose updates via pull requests.
README
organisation.md
your team's living memory — an mcp server backed by a git repo.
organisation.md turns any github repository into your organisation's persistent context layer. any mcp-compatible ai tool (claude code, cursor, etc.) can read, search, and propose updates to your team's shared context — without a database, without a hosted service, without leaving your github account.
how it works
- fork this repository into your github organisation.
- edit
organisation.mdwith your team's identity, projects, decisions, and preferences. - run the mcp server — your ai agents connect and read/write context naturally.
all changes go through pull requests — your team reviews and merges them. git provides version history, branching, and access control out of the box.
quickstart
1. generate a personal access token
go to github.com/settings/tokens and generate a classic token with the repo scope. copy the token — you'll need it in the next step.
2. run with npx (recommended)
export GITHUB_TOKEN=ghp_your_token_here
export GITHUB_OWNER=your-github-username-or-org
export GITHUB_REPO=organisation.md
npx @shashank-sn/organisation-md
that's it. the mcp server starts in stdio mode and your ai agent can connect.
3. or clone and run locally
git clone https://github.com/your-org/organisation.md.git
cd organisation.md
npm install
export GITHUB_TOKEN=ghp_your_token_here
export GITHUB_OWNER=your-github-username-or-org
export GITHUB_REPO=organisation.md
npx tsx src/server.ts
4. connect to your ai agent
the server speaks mcp over stdio. configure your ai tool to launch it:
claude code — add to your .mcp.json:
{
"mcpServers": {
"organisation.md": {
"command": "npx",
"args": ["@shashank-sn/organisation-md"],
"env": {
"GITHUB_TOKEN": "ghp_...",
"GITHUB_OWNER": "your-org",
"GITHUB_REPO": "organisation.md"
}
}
}
}
cursor / continue / other mcp hosts — point to the same command with env vars.
tools
| tool | description |
|---|---|
read_org |
read the full organisation.md file from the github repo |
read_section |
read a specific section by heading (e.g., "team", "decisions") |
update_section |
propose an update to a section — creates a pull request |
search_context |
search across organisation.md and all context/ files |
propose_change |
propose a change to any file in the repo — creates a pull request |
list_context_files |
list all files in the context/ directory |
import_file |
import a txt, md, or docx file into the knowledge base |
add_info |
add information via natural language — auto-detects the section |
remove_info |
remove information matching a description |
check_roles |
check git-based roles and codeowners |
check_permissions |
check if you can add/delete/approve/merge |
configure_codeowners |
update .github/codeowners for team access control |
report_bug |
auto-detect the area and file a github issue |
suggest_feature |
suggest a feature or improvement as a github issue |
repository structure
organisation.md/
├── organisation.md # canonical org context file
├── context/ # supporting context files
│ ├── README.md
│ ├── projects.md # detailed project info
│ ├── architecture.md # system architecture
│ └── people.md # team members and roles
├── src/ # mcp server source (typescript)
│ ├── server.ts # entry point
│ ├── github/ # octokit wrapper (files, git api)
│ ├── content/ # markdown parser and template
│ ├── tools/ # mcp tool implementations
│ └── resources/ # mcp resource templates
├── docs/ # documentation
│ ├── quickstart.md # full setup guide
│ ├── agent-prompt.md # prompt template for ai agents
│ └── example-flows.md # common workflows
├── scripts/
│ └── build-site.mjs # github pages build script
├── .github/workflows/
│ ├── ci.yml # ci: typecheck + test on push/pr
│ └── pages.yml # github pages deployment
├── strategy.md # product strategy
├── license # mit
└── README.md
environment variables
| variable | required | description |
|---|---|---|
github_token |
yes | github personal access token with repo scope |
github_owner |
yes | github username or organisation that owns the repo |
github_repo |
yes | repository name (defaults to organisation.md) |
customisation
- edit
organisation.md— replace the template content with your team's actual identity, projects, and decisions. - add
context/files — create additional markdown files for deeper context (architecture docs, runbooks, etc.). - fork and rename — fork the repo into your org and update the env vars to point at your fork.
development
npm install
npm run dev # development server with hot reload
npm test # run tests
npm run typecheck # typescript checking
npm run build # compile typescript
npm run build:site # build github pages site locally
license
mit — fork it, use it, ship it. see license.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.