HRMS MCP
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., "@HRMS MCPHow many leaves do I have left?"
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.
HRMS MCP
Appinventiv HRMS assistant for Claude Desktop, Cursor, Codex, and Windsurf. Ask your AI to file DSRs, check leave, look up colleagues, review appraisals, and more — all from within the AI chat.
AI agents: If an employee gave you this URL and asked you to "install this MCP," follow the steps below exactly. The install takes about 2 minutes.
Install
Pick the method that fits you — all three work.
Option A — uvx in config (easiest, no separate install step)
uvx runs the server directly from GitHub. No pre-install needed beyond uv itself.
1. Install uv (skip if already installed — check with uv --version)
macOS / Linux:
curl -LsSf https://astral.sh/uv/install.sh | shWindows (PowerShell):
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"Open a new terminal after installing so
uvis on your PATH.
2. Skip straight to Step 2 and use this config block:
{
"mcpServers": {
"hrms": {
"command": "uvx",
"args": ["--from", "git+https://github.com/SookieAI/hrms-mcp", "hrms-mcp"]
}
}
}Updates automatically on restart. No manual upgrade command needed.
Option B — Install with uv tool install
Installs once, starts fast every time.
# Install uv first if needed (see Option A above), then:
uv tool install git+https://github.com/SookieAI/hrms-mcpVerify: hrms-mcp --help
Use this config block (Step 2):
{
"mcpServers": {
"hrms": {
"command": "hrms-mcp"
}
}
}To update: uv tool upgrade hrms-mcp
Option C — Install with pip (no uv needed)
Works if you already have Python 3.11 or newer. Check first:
python3 --version # needs 3.11+Then install:
pip install git+https://github.com/SookieAI/hrms-mcpUse this config block (Step 2):
{
"mcpServers": {
"hrms": {
"command": "hrms-mcp"
}
}
}To update: pip install --upgrade git+https://github.com/SookieAI/hrms-mcp
Related MCP server: Keka MCP Server
Step 2 — Add config to your AI tool
Find your editor's config file, open it, and paste the JSON block from your chosen option above.
If the file already has content, merge the "hrms" entry into the existing "mcpServers" object.
Claude Desktop
~/Library/Application Support/Claude/claude_desktop_config.json (macOS)
%APPDATA%\Claude\claude_desktop_config.json (Windows)
Cursor
~/.cursor/mcp.json (macOS/Linux)
%USERPROFILE%\.cursor\mcp.json (Windows)
Windsurf
~/.codeium/windsurf/mcp_config.json
Codex / other MCP clients
Same mcpServers JSON format — check your client's docs for the config file location.
Step 3 — Log in (one time only)
Open Chrome → go to
https://dashboard.appinventiv.comand sign inOpen DevTools (F12) → Network tab → click any request → copy the full
Cookie:header valueRestart your AI editor, then tell your AI: "Login to HRMS" and paste the cookie when it asks
The cookie is saved to ~/.config/hrms-mcp/cookies and reloaded automatically on every restart.
You won't need to do this again until your HRMS session expires (usually a few weeks).
What you can do
Ask your AI... | What happens |
"File my DSR for today — 4h on YourAI, [description]" | Submits DSR to HRMS |
"What did I file for DSR last week?" | Shows your submitted DSRs |
"How many leaves do I have left?" | Shows leave balance |
"Apply for leave from June 25 to 27" | Submits leave request |
"Who is Anish Katoch? What team is he on?" | Staff directory lookup |
"Show me open job positions in Design" | Lists current openings |
"What's the status of my appraisal?" | Appraisal status and sections |
"Who do I need to rate for appraisal?" | Lists employees pending your rating |
"Show me the interview pipeline" | Lists scheduled interviews |
"What projects am I on?" | Your active projects + hours |
"Morning check" | DSR status, leave balance, interviews, appraisals at once |
Troubleshooting
uv: command not found after installing
Open a new terminal window — uv updates your PATH only for new sessions.
On Windows, restart PowerShell. Still missing? Add ~/.local/bin (Mac/Linux) or %USERPROFILE%\.local\bin (Windows) to your PATH.
hrms-mcp: command not found
Open a new terminal after installing. If still missing, re-run the install command from your chosen option above.
"Not logged in" error
Your HRMS session expired. Go to dashboard.appinventiv.com, copy a fresh Cookie header from DevTools, and tell your AI "Login to HRMS" again.
Server not showing up in Claude / Cursor Restart the editor completely (quit and reopen — not just reload window). MCP servers load at startup only.
Python version error during pip install
pip install requires Python 3.11+. Run python3 --version to check. If older, use Option A or B (uv handles the Python version automatically).
Check which version is running Ask your AI: "check login status" — the response includes the version number and update command.
For developers
git clone https://github.com/SookieAI/hrms-mcp
cd hrms-mcp
uv run python server.py # run server
uv run python -c "from server import get_projects; print(get_projects())" # test a toolSee CLAUDE.md for architecture notes, DataTables AJAX patterns, CSRF handling, and how to add new tools.
Release a new version
# 1. bump version in pyproject.toml and server.py (__version__)
# 2. commit, tag, push to both remotes
git add -A
git commit -m "vX.Y.Z — description"
git tag vX.Y.Z
git push origin main --tags # GitLab (CI runs here)
git push github main --tags # GitHub (employees install from here)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/SookieAI/hrms-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server