AI loops recurring tasks
A loop is a task that keeps coming back. It starts on its own, checks its result and stops when the goal is reached.
Prompt, skill, agent, loop: what is the difference?
- Prompt One instruction.“Write a friendly payment reminder.”
- Skill Written instructions for one task.Step by step, equally good every time.
- Agent An assistant with a goal.Uses several skills and asks you before important steps.
- LoopA recurring task.Runs on a plan, checks itself and stops at the goal.
A prompt is a single request, a skill a written work instruction, an agent like a colleague who knows several instructions. A loop is her fixed appointment: she does the task again and again and checks each time whether it is done.
For pros
- Prompt
- A single message in a chat. Good for one-off tasks.
- Skill
- A SKILL.md file in the open Agent Skills format. The assistant loads it when a request matches.
- Agent
- Instruction plus skills plus tools such as a mailbox or calendar, optionally with a trigger. Works through several steps; you approve.
- Loop
- Trigger plus goal plus check plus stop condition. Uses prompts, skills and agents; you approve steps with consequences.
By area
By job and industry
Managers 4Freelancers 3Small businesses 3Agencies 2Creators & media 2Office & admin 2Software teams 1Retail & online shops 1School & university 1
What is a loop?
A prompt is a single request. A skill is a set of work instructions, an agent an assistant with a goal. A loop runs these building blocks regularly, or until a check passes. Every good loop has five parts:
- Trigger
- What starts the loop: a schedule, an event such as a new email, or your request.
- Goal
- What should exist at the end, concrete enough to be checked.
- Check
- How the loop knows the goal is reached: a checklist, a source, a test or a person.
- Stop condition
- When it stops: goal reached or no more than the set number of rounds.
- Approval
- For steps with consequences, such as sending or paying, you decide yourself.
In the glossary: Loop engineeringRalph loopAgent loop
Three kinds of loops
Scheduled
Runs on a schedule, for example every morning or every Monday.
Goal-based
Repeats work and check until the result passes the check.
Event-based
Starts when something arrives, for example a new request.
Where the loop runs
Chat assistant with a schedule
ChatGPT offers scheduled tasks, Gemini offers schedules. Simple scheduled loops run there without other tools. Availability depends on plan and region.
No-code automation
In n8n, Make or Zapier you build trigger, AI step, check and approval as a workflow. n8n has its own evaluations for this.
n8n · Make · Zapier Agents
Coding agents
Claude Code has commands for interval loops, goal-based loops and scheduled routines. Codex has goals with a budget, Cursor has Automations on a schedule or event.
Vendors change features and plans often. The vendor's help page is what counts.
Sources: https://help.openai.com/en/articles/10291617-scheduled-tasks-in-chatgpt · https://support.google.com/gemini/answer/16316416 · https://docs.n8n.io/build/integrate-ai/test-and-improve-ai-workflows/run-quick-evaluations · https://code.claude.com/docs/en/scheduled-tasks · https://developers.openai.com/cookbook/examples/codex/using_goals_in_codex · https://cursor.com/docs/cloud-agent/automations
Keep an eye on costs
Every round uses compute. Subscriptions often have usage limits; with API access you pay per request. Limit the number of rounds and calculate beforehand.
Common mistakes
- No stop condition: the loop keeps running and uses compute without ever finishing.
- No real check: the assistant reports success although nobody looked.
- Too much freedom: the loop sends or changes something you wanted to see first.
Use, safety and notes
AI results can be wrong or incomplete. Check them before you use or send them. Not legal, tax, financial or medical advice. No warranty.
Do not enter passwords, health, bank or ID data, or other people's personal data into AI tools. Whatever you enter is processed by the provider under its own terms.
You may use and adapt our skills freely, including at work. Please do not republish them as your own collection. We give no warranty for AI results; check them yourself. Terms of use