Infinite Debug Loop: Let Cursor CLI Fix Until Tests Pass

Write a shell script that has cursor agent automatically fix code, run tests, and if they fail let the AI fix again — repeating until all tests pass. No need to sit there pressing Y/N. Hand the loop to the AI and go do something else — come back when the terminal shows "tests passed."

Core Concept

while tests don't pass:
    run tests → collect error output
    feed error output to cursor agent
    cursor agent fixes the code
    run tests again

It's a "run tests → fail → tell AI to fix → repeat" loop until green. Simple and direct, but effective.

Step 1: Check Prerequisites

cursor is installed and logged in (cursor --version to verify)
The project has a test command that runs in the terminal, such as:
- npm test
- pytest
- go test ./...
- cargo test
Tests return a non-zero exit code on failure (almost all test frameworks do this by default)

Step 2: The Simplest Script

Create debug-loop.sh:

#!/bin/bash
 
MAX_LOOPS=10        # max iterations to prevent truly infinite loops
TEST_CMD="npm test" # replace with your test command
loop=0
 
while [ $loop -lt $MAX_LOOPS ]; do
  echo ""
  echo "=== Loop $((loop + 1)) ==="
 
  # run tests and capture stdout + stderr
  TEST_OUTPUT=$(eval "$TEST_CMD" 2>&1)
  EXIT_CODE=$?
 
  if [ $EXIT_CODE -eq 0 ]; then
    echo "✅ Tests passed! Completed in $((loop + 1)) loop(s)."
    exit 0
  fi
 
  echo "❌ Tests failed, letting the AI fix..."
  echo "$TEST_OUTPUT" | tail -30  # print the last 30 lines for your reference
 
  # feed the error output to cursor agent
  cursor agent "Tests failed with the following errors. Please fix the code to make the tests pass:
 
$TEST_OUTPUT" --model gpt-codex-5.3-high --no-interactive
 
  loop=$((loop + 1))
done
 
echo "⚠️  Reached max loops ($MAX_LOOPS). Stopping. Please investigate manually."
exit 1

Step 3: Run the Script

chmod +x debug-loop.sh
./debug-loop.sh

Now you can make coffee, reply to messages, and come back when you see ✅ Tests passed! in the terminal. If it hits MAX_LOOPS without passing, it's time for human intervention.

Step 4: Customize for Your Project

Change the test command (line 4):

TEST_CMD="pytest -x"          # Python, -x stops on first failure
TEST_CMD="go test ./..."      # Go
TEST_CMD="cargo test"         # Rust
TEST_CMD="./vendor/bin/phpunit" # PHP

Adjust the max loops:

MAX_LOOPS=5   # conservative, stop early if not improving
MAX_LOOPS=20  # more complex problems get more chances

Give the AI more context (inside the cursor agent command):

cursor agent "Tests failed. Errors below.
The test framework is Jest and the language is TypeScript.
Do not modify test files — only change files under src/.
 
Error output:
$TEST_OUTPUT" --model gpt-codex-5.3-high --no-interactive

Enhanced Version: Keep Logs for Each Loop

#!/bin/bash
 
MAX_LOOPS=10
TEST_CMD="npm test"
LOG_DIR="debug-logs"
mkdir -p $LOG_DIR
 
loop=0
 
while [ $loop -lt $MAX_LOOPS ]; do
  echo ""
  echo "=== Loop $((loop + 1)) ==="
 
  LOG_FILE="$LOG_DIR/loop-$((loop + 1)).log"
  TEST_OUTPUT=$(eval "$TEST_CMD" 2>&1 | tee "$LOG_FILE")
  EXIT_CODE=${PIPESTATUS[0]}
 
  if [ $EXIT_CODE -eq 0 ]; then
    echo "✅ Tests passed! Completed in $((loop + 1)) loop(s)."
    exit 0
  fi
 
  echo "❌ Failed. Log saved to $LOG_FILE"
 
  cursor agent "Tests failed with the following errors. Please fix the code to make the tests pass:
 
$TEST_OUTPUT" --model gpt-codex-5.3-high --no-interactive
 
  loop=$((loop + 1))
done
 
echo "⚠️  Reached max loops. Stopping. Check the logs in $LOG_DIR/."
exit 1

Useful for reviewing "what went wrong in which loop" after the fact.

Safety Considerations

Risk	How to Mitigate
AI breaks more files	`git commit` first to save state, `git checkout .` to revert if needed
Too many loops waste quota	Set a reasonable `MAX_LOOPS` (5–15 recommended)
AI modifies test files	Explicitly say "do not modify test files" in the prompt
Different failures each loop	Keep logs (enhanced version) for post-analysis

Summary

Script structure: run tests → fail → feed to AI → repeat until pass or hit limit
Key parameters: TEST_CMD (test command), MAX_LOOPS (maximum iterations)
Be specific in your prompt: state the language, restrict which directories to modify, no touching test files
git commit first — easy to revert if things go wrong, and peace of mind

Pair with Worktree: run a debug loop on different branches simultaneously. See 06-cursor-cli-with-worktree.