# Prompt Caching — Agent Loops, Stop Conditions and Token Budgets

Source: https://www.geekswithgeeks.com/en/agent-loops/budget-caching

> Reuse a stable prompt prefix across turns to cut cost and latency.

## Pay less for the part that does not change

The start of every request (system prompt, tool definitions, early history) is identical turn after turn. **Prompt caching** lets the provider remember that prefix, so cache reads are billed at a lower rate and are faster. With the Claude API you mark a block with `cache_control`. Put stable content first and changing content last, because only an unchanged prefix can be reused.

## Marking a cacheable block

The marker goes on the last block of the stable prefix. Minimum cacheable sizes, lifetimes and prices depend on the model, so check current documentation.

```python
system = [{
    "type": "text",
    "text": LONG_INSTRUCTIONS,
    "cache_control": {"type": "ephemeral"},   # cache everything up to here
}]
reply = client.messages.create(
    model=MODEL, max_tokens=1024, system=system, tools=TOOLS, messages=messages)
```

## Do not edit the prefix

Inserting a timestamp or changing the order of tools at the top of the prompt breaks the cache for every following turn. Keep the early part byte-for-byte stable.

**Quiz:** Where should changing content go for caching to work well?

- [ ] At the very start of the prompt
- [x] After the stable prefix
- [ ] In the API key
- [ ] Nowhere

*Answer:* After the stable prefix. Only an identical prefix can be reused, so volatile parts must come after it.