deepseekplugins

DeepSeek Harness plugin

dsh-context-structurer

Structured context for DeepSeek Harness: split the conversation into typed sub-contexts (reasoning / tool calls / replies) and index them with a table of contents. 结构化 dsh

Tools & Capabilities
Jump to install

Source facts

Repository
kkkkkklze/dsh-context-structurer
Latest update
Aug 16, 2026
Category
Tools & Capabilities
GitHub stars
0
Format
plugin
Catalog evidence
Upstream dsh.bundle evidence
Evidence path
package.json#dsh.bundle
Checked against
0.1.0-rc.8
Upstream check date
2026-08-20

This evidence comes from the upstream catalog. This site has not installed, run, or security-reviewed the plugin.

Install

Start with a prompt that asks an agent to review the GitHub repository and source. Switch to the command if you want to install it yourself.

Copy this prompt into DSH, Codex, or another agent and ask it to review the GitHub repository and source first.

Do not install or run any commands yet. Read this plugin's GitHub repository, README, and relevant source code. Then answer the questions below clearly and directly so I can decide whether it fits my needs:

1. What is this plugin, and what problem does it solve?
2. Who is it for, and what are its typical use cases?
3. How is it used after installation? Include one minimal example.
4. What known limitations or privacy, security, compatibility, or maintenance risks does it have?
5. Give a clear recommendation: recommend, conditionally recommend, or do not recommend, with reasons.

Distinguish statements documented by the repository, inferences from source code, and unknowns. If evidence is insufficient, say so explicitly. Do not guess or simply repeat the README.

GitHub: https://github.com/kkkkkklze/dsh-context-structurer
Plugin: dsh-context-structurer
Author: kkkkkklze

Check the source files

Read the README and other files from this plugin directory before installing.

File explorer2 files
README.mdSource · read only

dsh-context-structurer

结构化 dsh 上下文:把会话上下文拆成类型化小上下文(思考 / 工具调用 / 正式回复),以「结构化上下文目录」为索引。两种模式:

  • auto(默认):注入 目录 + 分型小节(思考/工具/回复摘要),内容变化才重发。
  • release(按需放行,省 token):目录升级为带序号的全量条目索引,三层自动筛选按用户需要放行相关条目,未放行的条目不占 token。

release 模式:三层筛选

用户发消息 → step 1 内部:
  ① 确定性基线(零成本):自动放行 最近回复 + 最近工具调用(可配置)
  ② 内置筛查器:一次小模型调用(输入=目录+用户请求 → 输出=相关序号)
     —— 主模型无感,不用调任何工具
  ③ context_release 工具(兜底):主模型可随时补放行
→ step 1 直接载入:目录 + 放行条目,主模型一次请求就绪
  • 确定性基线(baselineEnabled,默认开):最近 1 条回复 + 最近 2 次工具调用自动放行——模型什么都不做也有最近上下文;
  • 内置筛查器(screenerEnabled,默认开):纯文本小调用(无工具、无副作用),超时 15s、abort 联动、输出严格校验(仅目录内整数序号、上限 20 个),任何失败静默降级为基线,绝不阻塞会话;默认复用会话当前 provider/model,可用 screenerProvider/screenerModel 指定;
  • 工具兜底:context_release({ids:[...]}) 或 all,主模型补放行;放行工具的调用不占序号、不列目录。

安全设计

筛查器调用:纯文本补全(不给工具)、超时上限、abort 联动、输出白名单校验、失败降级;每个用户输入最多一次调用;screenerEnabled: false 可整体关闭(退化为纯基线 + 工具)。

序号稳定性

序号按条目身份分配(用户输入 / 思考块 / 工具调用+结果 / 回复块),追加式日志下跨 step、跨进程稳定,不会因新消息入日志而漂移。

配置(profile cordis.patch.yml 里 context-structurer 行的 config)

- id: context-structurer
  name: dsh-context-structurer
  config:
    mode: release            # auto | release(默认 auto)
    language: zh             # zh | en
    baselineEnabled: true    # 确定性基线开关
    baselineReplies: 1       # 基线自动放行的最近回复数
    baselineTools: 2         # 基线自动放行的最近工具调用数
    screenerEnabled: true    # 内置筛查器开关
    screenerMaxIds: 20       # 筛查器最多接受的序号
    screenerTimeoutMs: 15000 # 筛查调用超时
    screenerProvider: ''     # 空=复用会话模型
    screenerModel: ''
    maxSectionBytes: 4096    # 目录/放行正文的截断上限

auto 模式独有:includeThinking / includeTools / includeReplies(默认 true)、maxThinkingItems(5) / maxToolItems(8) / maxReplyItems(5) / tocMaxInputChars(120)。

联动(两种模式一致)

不 suppress、不禁用任何东西:扁平运行时快照(@deepseek-ai/dsh-system-prompt)与工作区指令(dsh-agent-instructions)的消息原样保留,与平台 compaction 不冲突。

安装

dsh plugin --profile <name> add C:/Users/Administrator/dsh-context-structurer

验收

node verify.mjs                # 契约 + 引擎 + 集成(93 断言;集成层需依赖可解析)
pwsh -File verify-plugin.ps1   # 完整验收:93 断言 + pmv2 双启动(release 模式)+ SHA-256 一致性
Back to searchOpen GitHub repository

More related plugins

View category