skills.rest
hAcKlyc
hAcKlycCOMMUNITY
Explore all skills by this author →
GitHub

myagents-task-automation

Create and govern scheduled, recurring, and condition-triggered MyAgents tasks via CLI.

863|101|Updated Jan 23, 2026
Search Tags:#CLI#scheduling#Workflow#CRON#Task Automation#myagents#condition trigger
One-click install
npx skills add https://github.com/hAcKlyc/MyAgents --skill myagents-task-automation
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: myagents-task-automation
Source: https://github.com/hAcKlyc/MyAgents/tree/main/bundled-skills/myagents-task-automation
Command: npx skills add https://github.com/hAcKlyc/MyAgents --skill myagents-task-automation

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Work that must outlive the current chat turn—reminders, periodic reports, continuous monitoring—often gets lost in conversation history or requires manual re-triggering. This Skill gives the Agent a complete mental model of MyAgents Tasks so it can persist intent, schedule future execution, route work to the right session, and govern the full lifecycle through the myagents task CLI.

Core Features & Use Cases

  • Task creation with scheduling: Create one-time (scheduled), fixed-interval, or Cron-based recurring tasks with explicit timezones, session routing (single-session vs new-session), and end conditions such as deadlines, max executions, or AI-initiated exit.
  • Command Detector activation: Attach a local script that cheaply checks external conditions on each tick and only wakes the AI when a condition is hit, avoiding unnecessary model calls; includes a strict stdin/stdout JSON protocol, checkpointing, and isolated fixture testing.
  • Lifecycle governance: Run, stop, start, rerun, archive, delete, inspect execution history, reset checkpoints, and post cross-session task comments, all with authoritative JSON receipts.
  • Use Case: Ask the Agent to watch a CI pipeline every 5 minutes and only notify you when a build fails. The Agent writes a detector script, validates it with fixture tests, creates the recurring task with a trigger file, and reports the deployment receipt with next execution time.

Quick Start

Ask the Agent to create a MyAgents task that runs a daily report every morning at 9 AM in your local timezone.

Frequently Asked Questions about myagents-task-automation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a scheduled or recurring task in MyAgents?▼

Write the action into a task.md file, then run myagents task create-direct with an execution mode: scheduled for a one-time dispatchAt, or recurring with a Cron expression or intervalMinutes. Finish by reading the JSON receipt, then call task get and task run to enable scheduling.

How do I make a task only run when a condition is met?▼

Use a command Detector: write a self-contained script that reads a JSON invocation on stdin and outputs a quiet or activate decision on stdout. Validate it with trigger test fixtures, then pass the production trigger file via --trigger-file when creating the task.

When should I use a task instead of running the work immediately?▼

Use a task when work must persist beyond the current chat turn, trigger in the future, or be tracked and governed over time. Immediate one-shot work should just be executed directly, simple saved notes belong in Records, and continuous multi-turn work in one session belongs in Goal Mode.

Does a MyAgents task keep running when the app is closed?▼

No. Tasks only tick while the MyAgents app is running, including when it resides in the background. If the app fully exits or the OS sleeps, ticks are not produced and missed executions are not replayed individually.

Why did my command Detector task stop activating?▼

Detector failures such as non-zero exits, timeouts, or invalid JSON output do not wake the AI. Recurring tasks back off and become Blocked after three consecutive failures; inspect task get for Trigger health and lastError, fix the script, retest with trigger test, then resume.

Related Skills

Explore More Product & Management Skills→
deandumdeandum
3

idea-refine

Refine vague ideas into actionable task statements with MVP scope and validation criteria.

Community
Advanced
Arcanada-oneArcanada-one
11

reflecting

Review completed tasks, verify quality gates, and generate reflection artifacts.

Official
Advanced
pluginagentmarketplacepluginagentmarketplace
3

continuous-monitoring

Detect adversarial activity and model drift in production AI systems.

Community
Advanced
1thirteeng31thirteeng3

hermes-agent-spawning

Spawn Hermes Agent subprocesses for isolated long-running tasks.

Community
Intermediate