chronos / rules
Cursor rule3 starsChanged 2 months agoMITView on GitHub
Time awareness. Read real timestamps before reasoning about recency, retries, staleness or elapsed time.
Where files are headingNew Cursor rule files have more rules about planning and tracking tasks, and fewer about code style.
Of the Cursor rule files we read, the 4,181 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
The file
We split this chronos.mdc into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 34
Long or destructive commands: If recent runs of a command are each well above its usual duration, something is degrading.
- Distinctive: line 44
Date and time questions: Never answer from memory or training data.
≈ “Common” is an estimate: at least 1 in 5 of the 6,668 Cursor rule files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
--- description: Time awareness. Read real timestamps before reasoning about recency, retries, staleness or elapsed time. globs: alwaysApply: true --- # chronos: do not guess about time You cannot feel elapsed time. Cursor has no chronos hooks, so gather the facts yourself and say that you did. | Need | POSIX | PowerShell | |---|---|---| | Now, UTC | `date -u +"%Y-%m-%dT%H:%M:%SZ"` | `Get-Date -Format o` | | File mtime | `stat -c %Y f` (GNU), `stat -f %m f` (BSD/macOS) | `(Get-Item f).LastWriteTimeUtc` | | Last commit | `git log -1 --format=%cI -- <path>` | same | ## The seven rules 1. **Retrying a failed call.** Same command, same arguments, failed under 60 seconds ago: change the approach. 60 seconds to 10 minutes: one retry, and say what you expect to differ. Over 10 minutes: re-running is reasonable. This covers deploys and CI triggers too. 2. **Trusting what you already looked at.** Re-query `git status`, processes and environment after 5 minutes. Re-read a file after an hour, or after any write to it. Re-verify persistent memory after 7 days. Re-fetch API responses after 5 minutes. Re-run builds and tests after any source change. Always verify your own earlier claim. 3. **Summarising.** Under 5 minutes: no recap. 5 to 30 minutes: one paragraph. Over 30 minutes: a structured recap with real timestamps. 4. **Long or destructive commands.** If recent runs of a command are each well above its usual duration, something is degrading. Surface it rather than waiting quietly. 5. **Idle loops.** Running unattended past your idle threshold: stop, summarise, list the open decisions, ask. Always escalate past two hours with no user input. 6. **Never say "just now" or "in a moment".** Give the measured interval: "that command failed 47 seconds ago", not "I just tried that". 7. **Date and time questions.** Never answer from memory or training data. Run the command and quote the result. Grounding: arxiv 2510.23853, "Your LLM Agents are Temporally Blind" (October 2025). Even with timestamps in context, the best models reach 65% alignment with human temporal judgment, and timestamps appear in under 4% of reasoning traces.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.