MyPlayground
This server lets you query real compensation data from levels.fyi for tech roles. It's recommended to call get_level_mapping first to understand a company's internal leveling structure before querying offers.
get_level_mapping: Find how a company names its internal levels (e.g., 'L5', 'SDE II', 'E4') for a given job family (e.g., Software Engineer, Product Manager).get_recent_offers: Fetch the most recent salary offers for a given company, role, and level, optionally filtered by location — useful for understanding current market rates and total compensation trends.
Allows querying compensation data from levels.fyi.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@MyPlaygroundadd 15 and 27"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
levels-fyi-mcp
MCP Server for querying compensation data from levels.fyi.
What this server provides
Server name: LevelsFyi
Tools:
get_recent_offers(company_name: str, role: str, level: str, location: str = None) -> dict(Hit levels.fyi salary search API to find the recent offer data for given company, role, level, location)get_level_mapping(company_name: str, role: str = "Software Engineer") -> dict(Get the level mapping for a given job family at a company)
End User Installation
This section is for users who only want to install and run the MCP server in Cursor.
Requirements
Python
>=3.11
Installing the MCP
This MCP is hosted on Prefect Horizon, which is a low-code/no-code way to host MCP servers for LLMs to be able to use. The MCP URL is https://levels-fyi.fastmcp.app/mcp, so you should be able to use this server address to add to Codex, Claude Code, OpenAI SDK, Cursor, Gemini CLI.
For example, in Claude Code, you can reference this wiki for how to install the MCP server. You can run claude mcp add --transport http levels-mcp https://levels-fyi.fastmcp.app/mcp to install the MCP server.
If you're on Cursor, you can visit this link and follow the on-screen instructions
Developer Setup
This section is for contributors working on code changes.
Install dependencies
From the project root:
uv syncRun server locally
uv run python main.pyAdd or update dependencies
uv add <package>
uv syncQuick environment check
uv run python -c "import fastmcp; print('fastmcp ok')"Project structure
.
├── main.py
├── src/
├── pyproject.toml
└── README.mdTroubleshooting
Failed to spawn: fastmcp (os error 2)
Cause: Cursor cannot find the fastmcp executable.
Fix:
Prefer Cursor config with
uv run python main.py, orInstall global tool with
uv tool install fastmcpand ensure PATH includes the uv tool bin (commonly~/.local/bin).
Contributing
If you have a feature request, please create a GitHub issue to triage the requests
Feel free to fix any issues but please raise a Pull Request so that I can review it prior to deployment
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
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/karkir0003/levels-fyi-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server