DeepSeek Harness Plugin Hub

Publish and manage complete Harness Profiles. Discover Plugins for your next setup.

Explore

PluginsPresetsDocsNews

Community

Publish a pluginContactReport an issue

Resources

Plugin Hub on GitHubDeepSeek HarnessSystem statusPrivacy notice
© 2026 DeepSeek Harness Plugin HubPowered byPaxTech

Independent and unofficial. Not affiliated with, authorized by, or endorsed by DeepSeek.

Compaction Codex — DSH Plugin for DeepSeek Harness
← Plugins
C

dsh-compaction-codex

Compaction Codex

Codex-style window-swap compaction engine: deterministic pointer-template summaries with zero LLM calls, replacing the default LLM-summary compaction.

The plugin will be installed here. Keep web if you are unsure.

npx -y @deepseek-ai/dsh plugin --profile web add github:r-ayin/dsh-compaction-codex#1d407e8e2c201ee458674c5a3b7f13c93b145a43
READMECompatibilityVersions

Compatibility and provenance

Compaction Codex is published as dsh-compaction-codex and currently resolves to version 0.0.1. The Hub verifies its manifest and preserves the exact installation source for reproducible installs.

DSH compatibility
*
Runtime surfaces
any
Release source
github
Registry updated
10/15/2026

Versions

0.0.1stable
10/15/2026

Related plugins

Loading related plugins…

Latest
0.0.1
DSH
*
HMR
Process restart
Tree shaking
Safe tree shaking not declared
Unpacked size
Unavailable
Files
Unavailable
Surface
any
License
MIT
Source
github
GitHub
★ 0
Weekly downloads
0
Last push
10/15/2026
View source ↗
README badge

Click the badge to copy Markdown for your README.

Do you maintain this Plugin?Claim benefit · Priority security scan

Verify the GitHub repository declared in package.json to manage this listing. After you claim it, Hub will prioritize a security scan of the current version and publish the result when it passes.

Claim this Plugin →
Report an issue
DeepSeek Harness Plugin Hub
ProfilesPluginsCategoriesNewsDocsSign inManage Profiles
ProfilesPluginsCategoriesNewsDocsSign in

Related plugins

More verified plugins in memory-context.

Hindsight Coding Agents@vectorize-io/hindsight-coding-agentsReflect-only Hindsight long-term memory for coding agents (harness-pluggable: opencode, …), with automatic background ingestion (no setup CLI).Ab Memorydsh-ab-memory--- description: "The dsh-ab-memory package: a cross-session, file-backed agent memory with five tools (remember / recall / forget / load_skill / distill) and an always-on light injection." kind: "package-reference" ---Contextdsh-contextA DeepSeek Harness plugin for context insight and management, with context dashboard and context command, for understanding how the context is made of, and how it evolves.Memsearch Dsh@zilliz/memsearch-dshMemSearch plugin for DeepSeek Harness: shared markdown memory across agents, with capture, pre-step context injection, memory-recall skill, and a skill-candidate review panel.

README

dsh-compaction-codex

Codex 式换窗压缩引擎——彻底替换官方 dsh-compaction-basic 的「LLM 摘要式压缩」。

机制

  • 继承官方 BasicCompactionEngine,只覆写官方承诺的唯一子类钩子 summarize();
  • 摘要改为纯确定性指针模板(零 LLM 调用):归档声明 + 恢复通道指引(history_search / history_read / notes read)+ 最近 model 检查点笔记全文(读 ~/.dsh/context-budget/notes/<sid>.jsonl,只读)+ 被归档区首尾消息截断锚点;
  • SummaryResult 走官方契约的模板分支(无 llmStreamCall 字段,官方注释点名 "template, remote, or other summarizer"),转录中 compaction/summary 事件显示 provider: template, model: codex-rollover;
  • shadow 事务、replay 定价、工具配对平衡、并发锁、阈值自动触发(0.8×窗口,router-standard 为 0.55)、context-overflow 恢复、/compact(compactNow)全部原样继承。

装配

  • 源码:~/.dsh/plugins-src/dsh-compaction-codex/(本目录),构建:bash scripts/build.sh(链接 dsh-install 类型依赖 + tsc 编译 src → lib);
  • profile 链:~/.dsh/profiles/web/node_modules/dsh-compaction-codex 软链 + package.json link 依赖(防 pnpm install 剪掉);
  • 预设:6 个用户预设(code / code-guarded / code-lean / dshagent / designharness / router-standard)compaction 组中 compaction-basic 条目已替换为 compaction-codex(2026-09-29,各文件留有 .bak-20260929 备份);
  • 生效范围:仅新会话(预设热扫描;运行中会话保持旧代装配)。

约束与边缘

  • 官方 summarizeCompaction 要求摘要帧 token 数 < 被 shadow 内容 token 数;本模板上限约 5K 字符(约 1.5K tokens),对数千 tokens 以上的压缩区安全;极小压缩区会按官方逻辑抛错并留在日志(不会静默);
  • notes 文件缺失/损坏时自动降级(省略检查点段),不阻塞压缩事务;
  • 回滚:预设 yml 中把 name 改回 @deepseek-ai/dsh-compaction-basic(或恢复 .bak)即回官方 LLM 摘要,热生效(新会话)。

决策记录

  • 2026-09-29 用户拍板:方案 A(子类替换)+ 纯 Codex 式(永不 LLM 摘要,无混合开关);
  • 否决方案 B(从零实现抽象 CompactionEngine:需重写 shadow 持久化/锁/配对平衡,收益零风险高)与方案 C(拆压缩组:长会话必撞上下文硬上限)。