Cursor rule

1---
2description:
3globs:
4alwaysApply: false
5---
6# Session-Based Logging Rule
7 
8**Your Task: Implement structured logging so each application run writes to a separate file in the `logs/` directory.**
9 
10## File Existence Check
11 
12Verify logging directory and ignore file:
13 
14```bash
15ls . docs
16```
17 
18Create missing items if necessary before proceeding.
19 
20## I. Directory & Git Ignore
21 
221. Ensure a `logs/` directory exists at the project root. Create it if missing.
232. Confirm `logs` (or `logs/`) is listed in `.gitignore` so log files are never committed.
24 
25## II. Log File Per Run
26 
271. On startup, generate a timestamp `YYYY-MM-DD-HH-MM-SS`.
282. Create a new file `logs/<timestamp>.log` and append all log output to it.
293. Subsequent runs create new files, preserving previous logs for reference.
30 
31## III. Log Content
32 
33- Use a logger like `pino` or `winston`, or simple file writes if unavailable.
34- Record start and end of major operations, successful paths, handled errors, and unexpected failures.
35- Include relevant environment info such as CLI arguments and Node version.
36 
37## IV. Toggle Mechanism
38 
39- Logging is enabled by default.
40- Provide an environment variable `LOG_LEVEL=off` or a command-line flag `--no-log` to disable file logging.
41- When disabled, skip creating the log file and suppress log output.
42 
43## V. AI-Friendly Format
44 
45- Prefer structured lines (JSON or clearly delimited text) for easy ingestion by debugging tools and AI systems.
46- Each entry should include level, timestamp, and message.
47 
48**REMEMBER: Logs must capture both happy paths and errors. Maintain readability while providing enough detail for effective troubleshooting.**
49