Warning

GitHub Security Notice:
GitHub prohibits use of their servers for excessive automated bulk activity or any activity that places undue burden on their infrastructure. Please review:

• GitHub Acceptable Use Policies
• GitHub Copilot Terms

Use TaskSync responsibly and at your own risk. You are responsible for ensuring your usage complies with GitHub's terms of service.

TaskSync

Reduce premium AI requests and manage tasks seamlessly with human-in-the-loop workflows. TaskSync provides three options to integrate feedback loops into your AI-assisted development.

Choose Your Option

Option 1: TaskSync (VS Code Extension) - Recommended

tasksync-demo.mp4

A dedicated VS Code sidebar extension with smart prompt queue system. Setup instructions here: tasksync-chat folder.

Features:

• Smart Queue Mode - batch responses for AI agents
• Give new tasks/feedback using ask_user tool
• File/folder references with # autocomplete
• Image paste support (copilot will view your image)
• Tool call history with session tracking

Installation: Install from VS Code Marketplace or build from source with npx vsce package.

---

Option 2: TaskSync Prompt

demo.mp4

A terminal-based task agent protocol. Your coding agent actively requests tasks or feedback through terminal commands, executes tasks autonomously, and operates until you stop it.

Features:

• Human-in-the-loop workflow - provide feedback or new tasks
• Terminal-based agent interaction
• Autonomous operation - runs continuously
• Never terminates automatically

Getting Started:

1. Choose your preferred prompt TaskSync Prompt
2. Provide it as context to your AI IDE
3. Send a prompt to the agent
4. Agent executes python -c "import sys; sys.stdin.read()" and waits for input

For spec-driven development, use Specs-Tasksync instead.

Note: Use Ctrl+D (Linux/Mac) or Ctrl+Z + Enter (Windows) to signal end of input.

---

Option 3: TaskSync MCP Server

This is an MCP server that helps with feedback-oriented development workflows in AI-assisted development by letting users give feedback while the agent is working. It uses the get_feedback tool to collect your input from the feedback.md file in the workspace, which is sent back to the agent when you save. Setup instructions: TaskSync MCP Server

---

Best Practices (VS Code Copilot)

For GPT models, use TaskSync MCP or Extension.

Recommended settings for agent mode:

"chat.agent.maxRequests": 999

Enable "Auto Approve" in settings for uninterrupted agent operation. Keep sessions to 1-2 hours max to avoid hallucinations.

Discussions

The TaskSync community can be found on GitHub Discussions where you can ask questions, voice ideas, and share your prompts with other people. Contributions to TaskSync are welcome and highly appreciated.

Star History

⭐ Drop a star if you find this useful!