I tried SKILL.state in my coding agents and stopped after four days

说明与用途

作者在编码代理中测试了 SKILL.state，但在四个任务后停止实验，发现状态文件增加了开销，而它解决的问题在升级到 Opus 5.5 后基本消失。

问题

Does an explicit state file improve long‑task coding agents or add unnecessary overhead?

结果

The SKILL.state approach improved brief clarity but introduced schema failures, invented timestamps, and linter issues; after moving to Opus 5.5 the context problem diminished, making the state file overhead unjustified.

工作流

• 1. Create a JSON state file per task containing objective, phase, decisions, findings, and next action.
• 2. Update the state only through a CLI that validates patches against a schema.
• 3. Launch a fresh subagent for each step, providing it the skill, current state, and latest observation, with no chat history.

证据

SKILL.state swaps an agent's growing chat history for a state file and a fresh step each time.

来源

The briefs were good. Each subagent got a small, self-contained frame and needed no history.

来源

Patches that failed the schema.

来源

Invented timestamps.

来源

After that, the median peak context was 162K tokens, and only 2 of 48 units of work were compacted at all.

来源