dsh-compaction-instant
VCC-style instant, near-lossless deterministic compaction engine for the DeepSeek Harness — a drop-in replacement for @deepseek-ai/dsh-compaction-basic
16 results
VCC-style instant, near-lossless deterministic compaction engine for the DeepSeek Harness — a drop-in replacement for @deepseek-ai/dsh-compaction-basic
Context compression tool (context_compact): the agent writes the replacement checkpoint itself and hands it to the host compaction engine, which skips the LLM summarizer call; automatic compaction stays on the official engine.
Content-aware tool output compression for DeepSeek Harness.
DeepSeek Harness plugin: gzip /api responses — fixes history loading failures over slow links
Context compression skill for DeepSeek Harness -- intelligent conversation summarization
附件自动降级:DSH 图片准入(单边 ≤2000px / ≤3.5MB / ≤4000万像素)超限图片不再拒绝,用 sharp 自动缩小/重编码后入库,用户上传手机原图不再踩线。
High-fidelity, faithful, bilingual, recursive compaction backend for DeepSeek Harness — a drop-in upgrade over dsh-compaction-basic.
Windowed, reversible context management for DeepSeek Harness; replaces Basic compaction.
Manual budget-targeted context compression for DeepSeek Harness: the conversation model picks ranges, a cheap flash-tier route writes the checkpoint summaries.
In-process DeepSeek Harness context compression bundle
Token optimizer for DeepSeek Harness — condense your context, keep the essence
DSH plugin for AST context compression, test log filtering, and token budget guard
High-fidelity, faithful, bilingual, recursive compaction backend for DeepSeek Harness — a drop-in upgrade over dsh-compaction-basic.
Active Context Pruning (ACP) for the DeepSeek Harness — model-driven context management as a CompactionEngine backend.
Out-of-tree dsh web-performance plugin: compression and immutable caching for shell assets and client bundles, plus preload hints — additive, disposable, no dsh source changes required.
Token-efficient project reading with symbol/import indexing, PageRank topology, and byte-level slicing for DSH