Rikinshah787/dotpals

Overview

A desktop utility that monitors coding agent output. It reads test results from agent output, flags fake passes, and can send Claude Code back to fix failures. Includes an MCP server for other AI tools to query.

Best for

Best for
Developers who want visibility into coding agent test results.

Use cases

• Monitor test results from Claude Code, Codex, Cursor, or Gemini CLI
• Detect faked test passes in agent output
• Trigger Claude Code to fix broken tests via MCP

How to use

Install

npx dotpals@latest setup

Tools exposed

• agents_now
• test_status
• ready_to_merge
• risky_steps
• handoff_note
• check_my_work

Tested with

Claude Code, Cursor, Continue, VS Code, ChatGPT

Notes

A desktop utility that monitors coding agent output. It reads test results from agent output, flags fake passes, and can send Claude Code back to fix failures. Includes an MCP server for other AI tools to query.

40 stars on GitHub. Last updated 2026-10-08. Licensed MIT.

Use cases

• Monitor test results from Claude Code, Codex, Cursor, or Gemini CLI
• Detect faked test passes in agent output
• Trigger Claude Code to fix broken tests via MCP

Pros

• Works with multiple coding agents
• Catches dishonest test reporting
• Provides an MCP server for integration

Cons

• Small project with limited adoption (40 stars)
• Requires a desktop environment
• May need per-agent configuration

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.