learn-loops

An interactive lesson on agentic loops, where an AI agent repeats actions until it reaches a defined result. It teaches how to choose a loop pattern and add safety limits through questions and examples.

In plain words
What is it for?
Use it to learn loop design from a real task, understand why success must be binary and verifiable, and prepare to hand the task to a loop architect.
Why use it?
It helps you learn the reasoning behind automated agent workflows instead of only receiving a generated setup.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/fltman/loop-engineer/learn-loops
Clone the repo
git clone --depth 1 https://github.com/fltman/loop-engineer

Made for: Claude Code.

Per session 14 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 165 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00014 $0.00165
Opus 5 $0.00007 $0.00082
Sonnet 5 $0.00003 $0.00033
Haiku 4.5 $0.00001 $0.00016

Measured 2d ago against content hash 0c308bc6e11a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

learn-loops scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/learn-loops.md · 16 lines

What it actually says

Use the loop-teacher agent to teach loop engineering interactively.

Starting point: $ARGUMENTS

If a task was given, teach through that task. If not, ask the user for a real task they'd want a loop to do, then teach Socratically from there — the core shift (source code → agent → loop), the one rule (binary, verifiable exit condition), the decision tree, and the guardrails. One idea at a time; let the user reason it out. Offer to hand off to loop-architect once they can pick a pattern and state an exit condition on their own.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 16 lines · 14 tokens per session scan A 0c308bc6e11a

Subscribe to this mod's changes

learn-loops is a command published in the GitHub repository fltman/loop-engineer (35 stars, last pushed 2mo ago), licensed MIT. It adds 14 tokens to every session and 165 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.