---
title: "Agents hate friction: early thoughts on building for agents"
description: "Designing software for LLMs and agents requires more than extending human developer experience: a polished text-first interface can still produce an AX that is merely functional. The post frames an agent as a combination of model, harness, context, and skills, with behavior varying across providers, models, and harnesses; it focuses attention on commonly used combinations while testing beyond them. In a test setup, agents receive goals and different amounts of context, and model-harness combinations are evaluated by how they reason, which tools they call, how many steps they take, and how consistently they repeat a task. The resulting design questions include friction, context and token efficiency, familiarity with established tools, and the trade-off between fewer steps and higher context or cost; the central conclusion is that agents may abandon a product when friction persists, making testing necessary to expose that behavior."
---

# Agents hate friction: early thoughts on building for agents

[Clickhouse](https://yomu.fyi/company/clickhouse) · Jun 29, 2026

**Type:** Explainer

## Summary

Designing software for LLMs and agents requires more than extending human developer experience: a polished text-first interface can still produce an AX that is merely functional. The post frames an agent as a combination of model, harness, context, and skills, with behavior varying across providers, models, and harnesses; it focuses attention on commonly used combinations while testing beyond them. In a test setup, agents receive goals and different amounts of context, and model-harness combinations are evaluated by how they reason, which tools they call, how many steps they take, and how consistently they repeat a task. The resulting design questions include friction, context and token efficiency, familiarity with established tools, and the trade-off between fewer steps and higher context or cost; the central conclusion is that agents may abandon a product when friction persists, making testing necessary to expose that behavior.

## Context

Human and agentic developers both value text-based interfaces, discoverability, readable errors, sane defaults, and good documentation, but they use these interfaces differently. A strong human developer experience can therefore result in an agent experience that is only functional. Agents can discover, use, and choose products, and persistent friction may cause them to abandon one for an alternative.

## Approach / What changed

Treat the agent as the user and evaluate combinations of models, harnesses, context, and skills. The testing setup gives agents a goal with varying amounts of context, then examines their reasoning, tool calls, step counts, and variation between repeated runs. The post also considers context efficiency, token cost, and the benefits and risks of designing around patterns familiar from tools such as gh, kubectl, and git.

## Takeaways

- An agent’s behavior depends on more than its model: harnesses, injected context, and optional skills also shape how it uses a product.
- Repeated testing can measure variation between agent runs, including tool selection and call counts, and can reveal whether design choices improve consistency.
- Reducing friction involves trade-offs among context size, number of steps, token cost, and familiarity with established tool patterns.

**Tags:** [Developer Experience](https://yomu.fyi/topic/developer-experience), [LLMs](https://yomu.fyi/topic/llm), [Testing](https://yomu.fyi/topic/testing)

- Source: [Clickhouse](https://clickhouse.com/blog/agents-hate-friction)
- Source URL: https://clickhouse.com/blog/agents-hate-friction
- Ingested by Yomu: 2026-08-29T01:37:08.194Z

[Read original post](https://clickhouse.com/blog/agents-hate-friction)