Cline团队从Meta Muse agent系统提示中提取指令并应用到自身harness，使编码agent性能提升：token减少2.7倍，速度提升2倍，成本降低2.4倍
AI 摘要
Meta声称Muse Spark 1.2与Muse agent harness共同训练，Cline团队从中提取系统提示指令应用到自身harness。在修复真实bug任务中，修改后的harness相比原始版本token减少2.7倍（19.7M→7.2M），速度提升2倍（49min→24min），成本降低2.4倍（$7.69→$3.25）。这表明系统提示对agent性能影响巨大，提示工程可带来显著提升。 核心观点： 1. 提取指令强调信任源代码、重视边缘案例、修复前重现bug、持续验证完成。 2. 相同模型Muse Spark 1.2下，仅改变提示使token减少2.7倍、速度提升2倍、成本降低2.4倍。
推荐理由高信息密度，值得细读
原文
So we did a fun experiment: Meta claims Muse Spark 1.2 was co-trained with their Muse agent harness. So we extracted instructions from their system prompt and added them to the Cline harness.
TL;DR of this special prompting: - Trust source code over the user prompt, so read every call site and existing tests before starting the task - Weigh edge and error cases as heavily as the happy path - Always reproduce the bug before fixing - Don't trust the first passing test suite, and verify suspicious looking half-baked tests - Never stop at just editing, keep working until the change is verified complete.
We then asked this modified harness to fix a real bug from our repo, and compared the results to the original Cline agent harness.
Results: - Used 2.7x fewer tokens (19.7M → 7.2M) - Finished 2x faster (49min → 24min) - Cost 2.4x less ($7.69 → $3.25)
Same Muse Spark 1.2 model, same task, only the prompting changed. Incredible how much of a performance gain Meta was able to achieve training it on these special instructions!
金句
Same Muse Spark 1.2 model, same task, only the prompting changed.
讨论
暂无评论。