Timesheet Assistant MCP
Generates timesheets from activity data and automates submission to PSI Project Server (SharePoint-based systems) using browser automation. Works with Activity Collector MCP to fetch data from GitLab, GitHub, and Calendar services.
README
Timesheet Assistant MCP
An MCP (Model Context Protocol) server for generating timesheets and submitting them to PSI Project Server (SharePoint-based timesheet systems).
Features
- Timesheet Generation: Generate daily, weekly, monthly, or custom date range timesheets
- PSI Integration: Automated submission to PSI Project Server
- Activity Data Integration: Works with Activity Collector MCP for data fetching
- Smart Summarization: Automatically summarizes work descriptions to 255 characters
- Browser Automation: Uses Puppeteer for SharePoint authentication and form filling
- Task Management: Hierarchical task tree navigation for PSI
Prerequisites
⚠️ Important: This MCP requires Activity Collector MCP to fetch activity data from GitLab, GitHub, and Calendar services.
Installation
Via npx (Recommended)
npx timesheet-assistant-mcp
Via npm
npm install -g timesheet-assistant-mcp
From Source
git clone https://github.com/sharadmathuratthepsi/timesheet-mcp.git
cd timesheet-mcp
npm install
npm run build
Configuration
For Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"activity-collector": {
"command": "npx",
"args": ["activity-collector-mcp"]
},
"timesheet-assistant": {
"command": "npx",
"args": ["timesheet-assistant-mcp"]
}
}
}
For Claude Code (VS Code)
Add to ~/Library/Application Support/Code/User/mcp.json:
{
"servers": {
"activity-collector": {
"type": "stdio",
"command": "npx",
"args": ["activity-collector-mcp"]
},
"timesheet-assistant": {
"type": "stdio",
"command": "npx",
"args": ["timesheet-assistant-mcp"]
}
}
}
Configuration File
Create ~/.timesheet-assistant-mcp-config.json:
{
"psi": {
"url": "https://projectserver.thepsi.com/PWA/_layouts/15/pwa/Timesheet/MyTSSummary.aspx"
}
}
Usage
First Time Setup
- Configure PSI credentials:
Configure PSI with username: YOUR_USERNAME and password: YOUR_PASSWORD
Generating Timesheets
- Daily:
Generate timesheet for today - Weekly:
Generate timesheet for this week - Monthly:
Generate timesheet for November 2024 - Date Range:
Generate timesheet from 2024-12-01 to 2024-12-07
Submitting to PSI
- Generate timesheet (using Activity Collector MCP)
- Get PSI tasks (call once, reuse for all days):
Get PSI tasks for 2024-12-05 - Select task from the hierarchical tree
- Submit timesheet:
Submit to PSI for 2024-12-05 with task index 3 and description "..."
Available Tools (8)
Timesheet Generation (4 tools)
generate_timesheet- Generate monthly timesheetgenerate_weekly_timesheet- Generate weekly timesheetgenerate_daily_timesheet- Generate single day timesheetgenerate_date_range_timesheet- Generate custom date range timesheet
Note: These tools orchestrate the Activity Collector MCP to fetch data.
PSI Integration (4 tools)
configure_psi- Configure PSI credentialsget_psi_tasks- Get available tasks from PSI (call once per timesheet period)submit_to_psi- Submit timesheet entry to PSIinspect_psi_page- Debug PSI page structure
How It Works
Timesheet Generation Flow
- LLM calls timesheet generation tool (e.g.,
generate_weekly_timesheet) - Tool returns instructions for fetching data from Activity Collector MCP
- LLM uses Activity Collector tools (
fetch_gitlab_activity,fetch_github_activity,fetch_google_calendar_events) - LLM builds and formats the timesheet
PSI Submission Flow
- Call
get_psi_tasksonce to get available tasks - User selects a task by index
- Call
submit_to_psifor each day (reusing same task index) - Browser automation fills SharePoint form and saves
PSI Integration Details
- Uses Puppeteer for browser automation
- Supports HTTP Basic Auth for SharePoint
- Handles lazy-loaded accordion trees for task selection
- Smart grid cell detection for SharePoint JSGrid
- Automatic form filling with MouseEvent simulation
Security
- PSI credentials stored in
~/.timesheet-assistant-mcp-tokens.jsonwith restricted permissions (600) - Browser runs in non-headless mode by default (can see what's happening)
- All data remains local
Troubleshooting
Activity Collector Not Available
Error: fetch_gitlab_activity tool not found
Solution: Install and configure Activity Collector MCP
PSI Submission Fails
Error: Could not find hours cell
Solution: Run inspect_psi_page to check PSI page structure
Browser Crashes
Solution: Ensure Puppeteer dependencies are installed:
npm install puppeteer
License
MIT
Contributing
Issues and pull requests welcome!
Author
Sharad Mathur (sharad.mathur@thepsi.com)
Related Projects
- Activity Collector MCP - Required companion MCP for data fetching
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.
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.
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.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
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.