Plugins

The official billion-context plugin: a context-compression plugin for small context windows (a 100K context is enough), token savings (5x fewer tokens), and month-long single sessions (billions of tokens).

Model-driven context compression (Active Context Pruning) for DeepSeek Harness: the model decides when and what to compress.

Access the dsh web UI over LAN: HTTP/HTTPS/WS forwarding plus TLS (self-signed or custom certificates); adaptive Brotli/gzip compression for HTTP and permessage-deflate for WebSocket; half-open WebSocket probing keeps mobile clients from going stale after backgrounding; launch-token auto-injection lets LAN devices connect without copying the token; DNS-rebinding protection + loopback-only upstream allowlist.

Gzip compression for /api responses, fixing history-load timeouts over remote/low-bandwidth connections; skips SSE and already-encoded responses.

Automatic context compression for DeepSeek Harness: compresses tool outputs before they reach the model and keeps every lossy compression reversible via CCR retrieval tools.

Local-first code knowledge graph for coding agents: 10 MCP tools for layered context compression, DAG planning, incremental AST indexing, a cross-session skill flywheel, diagnostics, and graph artifact import/export.

China A-share financial data MCP server for the harness: 11 fin_data__* tools (quotes/K-lines/financials/calendar/special-data/announcements/EDB/reconcile/funds/indices) over THS free official REST (primary) + AKShare (fallback) + Wind (exclusive domains), with observable degradation, context-budget compression (result-side -72.2% measured), and an offline 10-year market-data lake; BYOK.

Dedicated model routes, tools, and system guidance for vision, compaction, reviews, subagents, titles, and image generation.

Forces context compression when running low-context llama.cpp models such as Qwen3.8-27B, improving model efficiency and usage smoothness, together with well-refined compaction prompts.

Access DSH from your phone or anywhere: Cloudflare Quick Tunnel public link + LAN HTTP/HTTPS direct access (auto self-signed cert), token auth, gzip compression, floating phone-icon panel with QR codes, and a NapCat QQ bot channel to fetch the link.

China A-share financial data MCP server for the harness: 11 fin_data__* tools (quotes/K-lines/financials/calendar/special-data/announcements/EDB/reconcile/funds/indices) over THS free official REST (primary) + AKShare (fallback) + Wind (exclusive domains), with observable degradation, context-budget compression (result-side -72.2% measured), and an offline 10-year market-data lake; BYOK.

Guarded context compaction for DeepSeek Harness (dsh): the LLM proposes, deterministic guards dispose — eager per-atom shrink (extract/summary/false under verbatim guards) + lazy reference-graph eviction (0-LLM) + byte-exact recall from an append-only log.

Deterministic context compression for dsh — zero LLM calls, rule-based pruning that preserves paths, commands, error traces, and open todos verbatim. Compresses ~80k tokens to a few hundred.

LLM-free lossless* compaction engine for DeepSeek Harness

Context Assembler plugin for DeepSeek Harness: topic-block context assembly, water-pressure topic splitting, tool trace/rewrite, reality recall injection and handoff planning.

A slim port of Headroom: compresses tool output, cuts up to 20% of context, without affecting the model's context cache or agent performance.

Network optimization for the DSH web UI: reduces transfer size with caching and compression to greatly speed up page loading, plus a connection-drop indicator and automatic reconnection. Ideal for users pursuing peak performance or using unstable networks.

Multi-platform message gateway and WeCom AI bot bridge with session context compression and smartsheet creation.

Forgetting-curve in-place dialogue compression for DeepSeek Harness: recent turns stay verbatim, middle turns are structurally dehydrated (zero-LLM), ancient turns fold into a bounded summary pair — unbounded conversations on a bounded memory budget.

人脑式上下文压缩逻辑，大幅减少长任务上下文占用以及token消耗

Shows real-time token and cost savings from the Headroom compression proxy in the DSH web UI: a Settings dashboard plus a persistent stats line under the composer.

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.

AST-based context compression, test log filtering, and token budget guard for DeepSeek Harness: strips redundant syntax, prunes noisy log lines, and enforces a token ceiling so the context window never overflows.

DSH 智能上下文压缩插件 —— 在有限窗口内管理认知资源，让 Agent 记住真正重要的信息。 Smart context compression for DSH — managing cognitive resources within limited windows, so your Agent remembers what truly matters.