agentleFS
bradbushSFAIbradbushSFAI/nolty/skills/cron-management/SKILL.md

cron-management

Skill1 starsChanged 6 months agoMITView on GitHub

Manage Nolty's scheduled crons. Triggers on "show/list crons", "add a cron", "enable/disable cron X", "delete cron Y", "run cron Z now", "cron status", "next cron". Reads cron-runner/cron-jobs.json and cron-runner/state/cron-state.json. Always validates schedules with croniter before writing.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. grandamenium/cortextos★ 106same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository5 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 2 are distinctive, and 5 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.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run python3 cron-runner/bin/cron-runner.py list and format as a Telegram-friendly table.

    line 23
  2. Distinctive:

    Remove the job. ALWAYS confirm first.

    line 57
  3. Distinctive:

    Only suggest these to the user — don't run them silently.

    line 85
  4. Very specific:

    ALWAYS read cron-jobs.json fresh before writing — the user may have edited it.

    line 89
  5. Very specific:

    Write atomically: write to .tmp then rename.

    line 90
  6. Very specific:

    Keep jobs array sorted by id after add/edit for stable diffs.

    line 92
  7. Very specific:

    Never edit cron-state.json — that's the runner's territory.

    line 93

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Cron Management

Nolty's interface for managing the cron-runner. The runner at cron-runner/bin/cron-runner.py is the dispatcher — this skill is the CRUD layer for cron-runner/cron-jobs.json.

Paths (adjust if your install location differs):

  • Config: <NOLTY_HOME>/cron-runner/cron-jobs.json
  • State: <NOLTY_HOME>/cron-runner/state/cron-state.json
  • Runner: <NOLTY_HOME>/cron-runner/bin/cron-runner.py

<NOLTY_HOME> is the path to your nolty/ checkout. Default: ~/Documents/CodingProjects/nolty.

Subcommands

Parse the user's intent into one of these actions. When ambiguous, ask. Never silently delete or fire a cron without confirmation.

list — "show me my crons" / "list crons" / "next cron"

Run python3 cron-runner/bin/cron-runner.py list and format as a Telegram-friendly table. Include: enabled flag, id, next fire time (in user's TZ), last fired, model/effort. If the user asks "next cron", just return the soonest-firing enabled job.

add — "add a cron that ..."

Conversational. Collect, in this order:

  1. id (kebab-case, must be unique; suggest from name)
  2. name (human label)
  3. schedule (cron expression — validate with croniter; show next 3 fire times for confirmation)
  4. slash_command (must start with /; verify the slash command file exists)
  5. model (haiku | sonnet | opus — default sonnet)
  6. effort_level (low | medium | high — default medium)
  7. category (info | marketing | ops | sales)
  8. max_lag_minutes — defaults by category: heartbeat=20, daily=60, weekly=1440, monthly=2880
  9. enabled (default false — the user enables explicitly after a dry-run)

Validate schedule:

from croniter import croniter
from datetime import datetime
from zoneinfo import ZoneInfo
now = datetime.now(ZoneInfo("America/Chicago"))  # use user's TZ
c = croniter(schedule, now)
[c.get_next(datetime) for _ in range(3)]

Then append to cron-jobs.json jobs array and confirm.

enable <id> / disable <id> — "disable the heartbeat cron"

Flip the enabled flag. Confirm with the user first if it's a high-visibility cron (heartbeat, morning-brief).

delete <id> — "delete the X cron"

Remove the job. ALWAYS confirm first.

edit <id> <field> <value> — "change daily-recap to 10pm"

Modify a single field. For schedule changes, validate with croniter and show new next-fire times before writing.

fire <id> / run <id> — "run X now"

Force-fire one job immediately: python3 cron-runner/bin/cron-runner.py fire <id>. This bypasses the schedule check, types the slash command into Nolty's tmux pane. Note: that would trigger you (Nolty) to dispatch a sub-agent. Alternative: the user can just type the slash command directly in Telegram.

status — "are crons healthy?"

Read cron-state.json. Report:

  • N enabled jobs
  • M fires in last 24h
  • Any jobs with last_status other than dispatched or dry_run
  • Most-recent fire and next upcoming

lookup <id> — internal

Return a job's full record. Used by Nolty's routing rule if she needs full context on a cron.

LaunchAgent control

  • Load (start the 15-min dispatcher): launchctl bootstrap gui/$(id -u) ~/Library/LaunchAgents/<your-label>.plist
  • Unload: launchctl bootout gui/$(id -u)/<your-label>
  • Check loaded: launchctl list | grep cron-runner

Only suggest these to the user — don't run them silently. Loading/unloading the dispatcher is a system-level state change they should know about.

Editing safety

  • ALWAYS read cron-jobs.json fresh before writing — the user may have edited it.
  • Write atomically: write to .tmp then rename.
  • Validate JSON parses after every write.
  • Keep jobs array sorted by id after add/edit for stable diffs.
  • Never edit cron-state.json — that's the runner's territory.

When the user's phrasing is ambiguous

  • "is the morning brief cron working?" → run status, then specifically lookup morning-brief last_fired_at and next fire.
  • "stop the heartbeat" → disable heartbeat (confirm).
  • "add a 9pm cleanup" → start add flow; ask "what does it do?" to derive slash_command.
  • "kill all crons" → confirm hard ("Disable ALL N crons? y/n"), then iterate disable per id.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.