← 返回列表

dsh-context-compression-improved

机制上与上下文压缩领域最响的两条公开路线同源：代码骨架闸门沿用 Headroom（Apache-2.0）的骨架化思路，其公开头条为「编程 agent 少 20% token、JSON 载荷少 60–95% token，答案不变」；估计器通道沿用 TokenPilot（arXiv:2606.17016）的缓存感知上下文管理思路，该论文报告长会话 agent 成本最高降低 60%。这两个数字都是来源方自己的口径，此处照引；本插件不自带 benchmark，不自称任何降幅。在此之上为 DeepSeek Harness 提供：在同一设置区选择压缩 Profile、调整 Auto Compact 触发水位并开关代码骨架压缩；基于 DeepSeek V4 官方 tokenizer 的精确计量与同修订计数校验；非支持模型自动 fail-open，保留原始工具结果。

Same-origin in mechanism with the two loudest lines in context compression. The code-skeleton gate follows the skeletonization approach of Headroom (Apache-2.0), whose published headline is 20% fewer tokens for coding agents and 60–95% fewer tokens for JSON, same answers. The estimator channel follows TokenPilot (arXiv:2606.17016), which reports up to 60% lower cost for long-session agents. Both figures are theirs, quoted as-is; this plugin ships no benchmark of its own and claims no reduction of its own. What it adds for DeepSeek Harness: choose a compression profile, set the Auto Compact trigger level and toggle code-skeleton compression from one settings section, with exact DeepSeek V4 tokenizer measurement, same-revision count verification, and fail-open behaviour that keeps the original tool results on unsupported models.