@deepseek-ai/dsh-client-ui-session-pin
Codex-style global pinned conversations UI plugin for DeepSeek Harness
109 results
Codex-style global pinned conversations UI plugin for DeepSeek Harness
Deterministic two-level memory injection for DeepSeek Harness: global and project MEMORY.md plus today's notes, auto-loaded into every new session.
DSH plugin that injects an editable section into the global system prompt, with a settings page to customize it.
Granular settings platform: one Granular Settings page in the settings dialog with Workspace/Session/Plugin tabs. Three scopes (workspace, session, global) registered by plugins, namespaced storage, doorbell-only push.
Pure chat mode for DeepSeek Harness: hide every global tool so the agent only talks. Toggle per session with /pure-chat on|off.
EN: Settings and default-route layer for official DeepSeek Harness subagents: session/role/global model routing with a composer selector. ZH: 面向 DeepSeek Harness Web 官方子代理的设置与默认路由层:会话/角色/全局模型路由与输入框选择器。
DeepSeek Harness global account-balance card: live session/job status plus GLM / KIMI quota windows and OpenRouter / DeepSeek balances, pinned to the shell overlay bottom-right
Render local image files inline in the DSH conversation via a global show_image tool.
DSH KITT — Spanish-first voice for the DeepSeek Harness web UI: hands-free conversation, push-to-talk (Whisper), read-aloud replies, and a floating overlay window with global hotkeys. · DSH KITT 语音:为 DeepSeek Harness 打造的语音插件——免提对话、按键说话(Whisper)、朗读回复,以及带全局快捷键的悬浮窗;界面支持中文、英文和西班牙语。
Resident-index + frecency file search for DeepSeek Harness — shadows the built-in grep/glob tools. · DSH 常驻索引 + frecency 文件搜索,同名覆盖内置 grep/glob。
Glob-activated rule prompts for DeepSeek Harness (DSH): when the agent reads or edits a matching file, the rule activates and is injected into the conversation (Claude Code rules.md / # Path: style).
Strict gate: turns strict_check and the failure journal from tools the model may call into policy the harness enforces — repeated-failure intervention, a glob-protected critical-path gate, and an automatic post-write check.
Permission matrix for DeepSeek Harness: 3 sandbox modes x 4 approval strategies = 9 selectable presets, with a global default, a separate LLM-robot default, four-tier risk policies, password-gated approval for high-risk operations, audit log and git checkpoint.