dsh-token-optimizer
Deterministic, recoverable tool-result compression and cache-aware compaction for DeepSeek Harness
52 results
Deterministic, recoverable tool-result compression and cache-aware compaction for DeepSeek Harness
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.
Improved fork of dsh-context-compression-selector with an orthogonal code-skeleton compression gate
In-process DeepSeek Harness context compression bundle
Active Context Pruning (ACP) for the DeepSeek Harness — model-driven context management as a CompactionEngine backend.
Pure-incremental directive-driven compaction for DeepSeek Harness: /compact-directive <requirement> summarizes the session middle per your requirement, /trim-directive <requirement> trims the whole conversation per your requirement
Request image optimization, intelligent retention, and observability for DeepSeek Harness (DSH)
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.
Image compression/conversion tools for DeepSeek Harness via api.compressio.cc MCP
AgentFrame compaction backend for DeepSeek Harness: semantic + physical KV compression (28.4x) replacing LLM summarization
DSH plugin: compress verbose voice-dictation text into token-efficient prompts, fully local.
Context compression for small-context models on the DeepSeek Harness Web GUI: one click condenses the whole conversation into a few sentences, opens a fresh session in the same workspace, and re-injects the summary as the first message so a small model keeps working with a clean, compact context.
Token optimizer for DeepSeek Harness — condense your context, keep the essence
PDF compression tools for DeepSeek Harness via api.pdfslim.app MCP
Custom wallpaper engine for the dsh web GUI: upload an image (client-side WebP/JPEG compression), frosted-glass blur, pane opacity, and auto font-color linkage.
Token-efficient project reading with symbol/import indexing, PageRank topology, and byte-level slicing for DSH
High-fidelity, faithful, bilingual, recursive compaction backend for DeepSeek Harness — a drop-in upgrade over dsh-compaction-basic.
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。
跨会话持久记忆压缩系统:生命周期 hooks 自动捕获工具观察、生成语义摘要、3 层检索工作流(search→timeline→get_observations,~10x token 节省)、渐进披露、私有标签、混合搜索。受 thedotmack/claude-mem(86k★ Apache-2.0)启发。
JS/CSS/HTML 代码压缩
把长文本渲染成图片发送来压缩 LLM 输入 token(DeepSeek 视觉模型、每图 384 token 封顶)— text-as-image token compression for DeepSeek Harness
Context compression skill for DeepSeek Harness -- intelligent conversation summarization
DSH plugin for AST context compression, test log filtering, and token budget guard