Examples

AI loop examples

The best AI loop examples are repetitive work patterns with clear evidence and a stop condition: code repair, source QA, browser testing, support follow-up, and pre-publish checks.

Explore by category ↓REAL WORK. REUSABLE PATTERNS.
01

Current library snapshot

68total loop patterns
10categories
21engineering loops
10operating domains

New to the idea? Start with what an AI loop is.

02

All examples

Engineering

21 LOOPS
01Claude Code Repo Readiness

You are about to let a coding agent do meaningful work in a repo and want fewer dumb mistakes.

02Project Docs Freshness

Code changed and the docs are probably now lying by omission.

03Test and Logging Coverage

Important flows are hard to debug or verify, especially auth, payments, data mutation, external APIs, jobs, and error boundaries.

04CI Optimization

The team is waiting on CI long enough that it changes behavior.

05Fresh Clone Onboarding

You suspect setup instructions only work for people who already know the system.

06Test Flake Stabilizer

CI fails differently across comparable runs.

07Cold Load Trim

A web app ships too much before showing anything useful.

08Architecture Rubric Refactor

A refactor matters, but “make architecture better” is too vague.

09API Contract Drift

Backend, frontend, docs, and SDKs may describe different realities.

10Acceptance Scenario Lockstep

A coding agent is about to implement a feature where success could be interpreted three different ways.

11Reference Oracle Implementation

You need a coding agent to implement complex behavior that can be checked against a browser, official library, legacy system, API, compiler, or production output.

12Agent Instructions After-Action

A coding agent just learned something about the repo, test harness, edge case, command, or convention that future sessions should not relearn the expensive way.

13Spec to Task Shards

The request is too large or vague for a single coding-agent prompt but not large enough to justify a full project plan ceremony.

14Behavior Ladder TDD

A pricing rule, validator, parser, permissions rule, workflow state machine, or other logic-heavy feature could balloon if implemented all at once.

15Trace-First Debugging

A failure has logs, screenshots, stack traces, recent commits, or user reports, but the cause is not obvious yet.

16Parallel Agent Worktree Sweep

You have multiple independent coding tasks and the human bottleneck is orchestration, not implementation speed.

17Completion Promise Loop

A coding agent keeps declaring work done after the easy part, or a feature has multiple acceptance criteria that must be proven before handoff.

18Feature Flag Retirement Sweep

Feature flags, experiments, or kill switches have outlived their rollout and are making code paths harder to reason about.

19Schema Migration Dry Run

A schema change, backfill, cleanup script, or bulk edit could corrupt data, lock tables, or surprise users if run directly in production.

20Webhook Replay Harness

An integration bug depends on third-party webhook payloads, retry timing, signatures, or idempotency behavior that is hard to reproduce from normal tests.

21Bug Report to Repro Kit

A bug report has real signal but is missing the exact steps, inputs, logs, screenshots, account state, or acceptance test an engineer needs to act without guessing.

03

How to choose

If the work is…Start with…
Code or repo maintenanceCoding agent loops
Fact checking or researchPre-Publish Source Check or Research to Artifact
Quality assuranceVerification loops
Recurring personal follow-upCalendar Promise Proof

Prefer loops with a deterministic verification loop, or filter every loop in the loop library. Designing your own? See loop engineering.

04

FAQ

What are examples of AI loops?

Examples of AI loops in this library are repeated work patterns with clear evidence and a stop condition: code repair, source checks, browser testing, support follow-up, and pre-publish checks. Each row names the job the loop is for.

How do I choose an AI loop?

Match the job. Start with a coding loop for repo work, a source check for research, a verification loop for quality, or a follow-up loop for recurring personal tasks. Prefer a loop whose verifier and stop condition you can actually run.

Are these AI loops free?

Yes. Every loop in the library is free to read and copy. The site does not run the task. An AI provider may still charge for the usage on your account.

Design the systemFind your next loop