Agent 連携 · DeepSeek Harness プラグイン — 235 ページ

Clawd the pixel pet, living in the DeepSeek Harness Web GUI: reacts to session and agent state, draggable, themable.

Clawd on Desk bridge for DeepSeek Harness, extended: context usage, approval/juggling/compaction signals, balance alerts.

A second brain for your DeepSeek Harness agent — strong-model review before risky tool calls, plus 70 models from one wallet. No accounts, no API keys.

Model-autonomous context compaction for DeepSeek Harness: mind_map surveys the agent's own context surface (seqs, token weights, clear-boundary markers and the clear-mind playbook), clear_mind replaces a chosen span with a model-written checkpoint through the platform's native compaction transaction, and a proactive reminder nudges the model when the context grows long. dsh-clear-mind：让模型自主压缩自身上下文——mind_map 俯瞰上下文表面（seq/token/清理边界标记 + 清理手册），clear_mind 把一段对话历史替换成模型自写的检查点（走平台原生 compaction 事务，GUI 可展开审计），上下文涨长或单轮步数过多时主动提醒模型清理。

让任何软件 agent 化：7 阶段 CLI 封装流水线（分析→设计→实现→测试→发布）、真实后端集成、JSON 输出、REPL 会话、refine 增量补全。受 HKUDS/CLI-Anything（48k★ Apache-2.0）启发。

DSH 环境与能力：在「设置」里列出这个 Agent 能操作什么 —— 工具注册表、技能、MCP 服务器，以及你自己安装的命令行工具与运行时；只读、容错，系统目录不列出

把 DeepSeek Harness 会话变成终端可交互的 CLI:终端直接与 agent 对话、切换会话/模型/模式/权限模式、审批桥接、codex 式工作区。

Installable DeepSeek Harness Web UI for replaying a completed session step

Attach files and folders in the DSH composer: clipboard paste, native file/folder picker, @-mention path search; folders are converted to paged Markdown for the agent to read page by page

Context firewall for DeepSeek Harness: detects credentials in tool results and replaces them with opaque placeholders before they reach the model — the secret never enters context, the task keeps going.

SQL 优化模式：EXPLAIN 分析、索引策略、N+1 解决、查询改写。受 wshobson/agents（38k★ MIT）启发。

SQL 优化模式：EXPLAIN 分析、索引策略、N+1 解决、查询改写。受 wshobson/agents（38k★ MIT）启发。

China-network bootstrap for DeepSeek Harness (dsh): connectivity probes to npm/GitHub/HuggingFace/mirrors, local proxy detection, mirror recommendations, and a generated PowerShell/bash bootstrap script. Read-only by default; apply is explicit. CLI + agent-callable cn_boot tool.

Cross-session memory plugin for DeepSeek Harness (dsh): file-first markdown memories, recall injection, agent memory tools

代码研读教练（Code Reading Coach）— 交互式引导研读论文对应开源代码的 Agent 预设插件。五段研读法：锚定论文主张 → 地形扫描 → 入口追踪 → 核心映射 → 闭环验证。

Pre-write reuse firewall for DeepSeek Harness: before the agent writes a new helper/service, surface the existing implementations that already cover that intent. Deterministic retrieval (no LLM) backed by the Auto_code_audit capability channel.

五轴代码评审：正确性/可读性/架构/安全/性能 + 质量门槛，合并前必审。受 addyosmani/agent-skills（88k★ MIT）启发。

五轴代码评审：正确性/可读性/架构/安全/性能 + 质量门槛，合并前必审。受 addyosmani/agent-skills（88k★ MIT）启发。

DeepSeek Harness 插件：为 Agent 增加 code_scan 工具，用 semgrep 扫描代码并输出按文件/行号/严重级别分组的中文报告

代码简化：删冗余、降复杂度、去不必要的抽象，让代码更简单可维护。受 addyosmani/agent-skills（88k★ MIT）启发。

代码简化：删冗余、降复杂度、去不必要的抽象，让代码更简单可维护。受 addyosmani/agent-skills（88k★ MIT）启发。

SQL 优化模式：EXPLAIN 分析、索引策略、N+1 解决、查询改写。受 wshobson/agents（38k★ MIT）启发。

SQL 优化模式：EXPLAIN 分析、索引策略、N+1 解决、查询改写。受 wshobson/agents（38k★ MIT）启发。

SQL 优化模式：EXPLAIN 分析、索引策略、N+1 解决、查询改写。受 wshobson/agents（38k★ MIT）启发。