Context trimmer: lists the current session's context entries with role, preview and ≈token estimate (cl100k), highlights over-threshold items, and removes selected entries exactly without LLM summarization.
Install
# from npm (prebuilt)
dsh plugin --profile web add @weibaohui/context-razor
# from GitHub (first run asks for allowBuilds approval — follow the hint, retry)
dsh plugin --profile web add github:weibaohui/context-razor
Any plugin you install runs third-party code with your own permissions — it can read your files, use your credentials, and reach the network, and tool approvals don’t sandbox it. GitHub-sourced plugins also run build scripts at install time — pnpm blocks those until you allow them, so an install can stop with ERR_PNPM_GIT_DEP_PREPARE_NOT_ALLOWED or ERR_PNPM_IGNORED_BUILDS; dsh prints the exact key to add under allowBuilds in your profile’s pnpm-workspace.yaml, and the install works on the next run. Allowing a build is a trust decision: only install sources you trust, and pin a commit (github:owner/repo#sha).
README
This plugin publishes its README in Chinese only.
上下文剃刀：把当前会话上下文逐条列出来——角色、预览、≈token（cl100k_base 估算，与技能市场同词表）——超阈值标红，勾选后精确裁剪。压缩不知道裁了什么，剃刀让你自己挑。
核心功能
- 逐条列出：当前会话模型可见的每条上下文（用户/助手/工具结果）按序排列，带预览、字符数与 ≈token 估算，点 ⋯ 查看全文与真实 usage
- 占比标注：每条的 token 徽章同时显示其占全部上下文的百分比（如
≈5728 · 15%），一眼看出谁是上下文大头 - 类型筛选：按类型一键过滤——用户 / 注入 / 助手，以及各工具名（bash、skill、read、edit…按会话实际出现的动态生成）。多选 toggle，与 token 量级筛选正交叠加
- 超阈值标红：可调阈值（默认 500 token，本地记忆），超限条目红色高亮；「只看超阈值」过滤 + 「token 高→低」排序，大头一眼可见
- 精确裁剪：勾选任意条目一键删除——连续段自动合并处理。删除走宿主 surface replace 协议（与官方 compaction 同款机制），不经 LLM 总结，删了什么、删了多少完全由你决定。每行尾部带 ✕ 快捷按钮，单条直接删与勾选后多选批量删走同一确认弹窗
- 不重排：本插件不会改变上下文条目顺序，只会剔除选中条目；「token 高→低」排序仅影响页面显示，模型可见顺序始终由会话 surface 决定
- 安全护栏：会话运行中禁止裁剪（等当前回合结束）；每次替换附带一条 notice 标记消息，模型知道这段历史被你移除了，需要时会主动向你确认
- 非破坏：append-only 日志保留全部痕迹（宿主机制如此）。「删除」= 从模型视野与界面投影中移除
安装
dsh plugin --profile web add @weibaohui/context-razor -w
装完重启 dsh web 即生效。入口：打开会话 → 顶部「上下文剃刀」标签（Hermes Loop 右侧），自动锁定当前会话。
删除的实现方式
dsh 会话是 append-only 事件日志（深冻结 + zstd 校验），物理删除不存在也不必要。本插件的裁剪 = 仿官方 dsh-compaction-tool-result-pruner 协议：对每个连续段追加一条 compaction/prune 计价事件 + 一条带 surfaceOp: {op:'replace'} 的 notice 标记消息（sourceEventSeqs 覆盖全部被影子化节点）。被删节点从此不进 deriveMessages()——模型看不到、UI 不再渲染、token 不再计入上下文，而日志留痕可审计。
联系我 :飞书群
版本兼容性
本插件与 DeepSeek Harness（@deepseek-ai/dsh）的版本对应关系：
| 插件版本 | 适配 dsh 版本 | 备注 |
|---|---|---|
| 0.4.5 | 0.1.7-rc.2 | 当前版本，已在 @deepseek-ai/dsh@0.1.7-rc.2 下验证运行 |
发版约定：每次发布新版本时，请在上表追加一行，记录该插件版本实际验证所用的
@deepseek-ai/dsh版本。package.json的engines.dsh声明最低支持版本；本表记录实际验证版本，二者配合使用。
Links
More in this category
vectorize-io/hindsight#coding-agents★ 52424
Hindsight, agent memory that learns: long-term project memory with auto recall and retain, knowledge pages, deep reflection, and per-repo memory banks.
volcengine/OpenViking#examples/dsh-memory-plugin★ 40541
OpenViking memory and context bundle for DeepSeek Harness: pre-step auto-recall and profile injection, session capture, `viking://` URI guarding, and recall/write memory tools backed by an OpenViking server.
agentscope-ai/ReMe#dsh★ 3667
Connects DeepSeek Harness to ReMe's local-first, self-evolving personal knowledge base: automatically captures completed main-agent conversations as user-owned Markdown memory, searches conversations and source material through reme_search with BM25, optional embeddings, and wikilink expansion, and schedules daily memory consolidation.
zilliztech/memsearch#MemSearch★ 2770
Shared Markdown memory for DSH and other coding agents, with automatic capture, pre-step context injection, searchable recall, and memory-to-skill self-evolution through a review panel.
vshulcz/deja-vu#extensions/dsh★ 1275
Reads the session files thirty-three other coding agents on this machine already wrote — Claude Code, Codex, Cursor, VS Code Copilot Chat, opencode, OpenClaw, Hermes, Kimi, Cline, Zed and more — including sessions from before it was installed: six tools (deja_recall, deja_session, deja_blame, deja_fix, deja_how, deja_remember), a /deja command, and optional automatic recall added to the runtime context. Local BM25 index, no LLM, no embeddings, no network (dsh plugin --profile web add dsh-deja).
adoresever/graph-memory★ 658
Traceable, searchable cross-session memory for DeepSeek Harness — conversation knowledge as typed graph nodes (TASK/SKILL/EVENT) and typed edges.
Community comments
Comments are public GitHub Discussions. Loading them connects to GitHub and Giscus; a GitHub account is required to post.