228 results
Weak-network adaptor for the DeepSeek Harness: transparent long-backoff model-stream retries (10-min cap), local response replay cache, heartbeat auto-reconnect, degraded-mode token economy, and a bilingual (zh/en) settings page. / 弱网适配插件：模型流长退避重试、本地响应缓存、心跳自动重连（默认 10 分钟封顶）、降级省 token、双语设置页。
Text-to-speech for dsh web: speaks each assistant reply out loud using Edge TTS neural voices (host bridge + browser player)
DSH regex tool: test/find/replace over text plus static pattern explanation, zero-dependency with ReDoS guards
DSH Web plugin that proposes concise next user messages above the composer and inserts a selected reply into the draft without sending it.
ChatVoice — free voice input + AI reply read-aloud for DeepSeek Harness (dsh). Zero config, zero cost, no API key, built on the browser's native Web Speech API. 给 DSH 装上「免费、免 API key、开箱即用」的语音输入 + 回复朗读闭环（中文优先）。
Expose a DeepSeek Harness agent over the Agent2Agent (A2A) protocol: agent card, JSON-RPC endpoint, LLM-powered replies, and a live status card.
DeepSeek Harness plugin: per-edit approval with a red/green line diff before write/edit/str_replace_editor — approve once or reject (Claude Code behavior).
DSH web plugin: a yellow one-line cache-miss notice under the first assistant reply of each turn, when that turn's first request rebuilt the prompt cache.
Per-message delete for DeepSeek Harness — a trash icon beside the copy button on every chat message that removes that message from the derived model context (and the visible transcript) via the official surface-replace contract
Host-neutral durable DAG workflows for Agents, Tools, Skills, MCP, triggers, replay, and visual Canvas
DSH web plugin: better-sidebar Git Remotes tab for fetch / ff-only pull / confirmed push. Does not replace the built-in Git stage/commit tab.
Diff cards in the DeepSeek Harness web chat for recognizable file mutations — MCP filesystem calls (edit_file, write_file), built-in edit/write, and file-mutating bash commands (heredoc replaces, sed -i, cat/tee/redirect writes).
Microsoft SkillOpt-Sleep integration for DeepSeek Harness: give your dsh agent a nightly sleep cycle that harvests past sessions, replays recurring tasks, and consolidates validated skills behind a held-out gate.
Web chat entry for DeepSeek Harness (dsh plugin): a sidebar button (right below New Session) that opens chat.deepseek.com in a dedicated desktop window when running under dsh-desktop (Wails bridge), falling back to a NEW browser tab in plain web / official browser mode (never replacing the current harness page)
DingTalk IM bot channel for DeepSeek Harness: each chat drives its own agent, replies and thinking come back as messages over the Stream-mode long connection.
Export a minimal, secret-scrubbed, replayable problem bundle for DeepSeek Harness via the /repro command.
DSH web plugin: a VS Code-style code editor view (file tree, Monaco, ripgrep search/replace, Markdown preview, git diff).
Reasonix-style cache-aware compaction backend for DeepSeek Harness (DSH). Replaces/enhances compaction-basic with compact_ratio, one structured summary checkpoint, and a stable recent tail.
DSH web plugin: 斗地主 — play Dou Dizhu in the DeepSeek Harness GUI while waiting for the agent to reply.
Windows minimal agent preset for DeepSeek Harness: a one-line fixed persona, gitbash + str_replace_editor + web_search, no runtime context, no compaction. Installs the preset into the user's agent-presets root.
Declare per-model reasoning efforts, image input, and openai-completions dialect on hand-declared llm-pi-ai routes. Writes the official namespace; does not intercept llm/stream or replace the Models page.
Multi-channel message bridge for DeepSeek Harness (DSH): route WeChat (ClawBot/iLink), QQ, and Feishu (Lark) chat messages to a DSH agent and relay its replies back. 多通道桥接：微信 ClawBot / QQ / 飞书消息接入 DSH agent 并回传回复。
Replay real DeepSeek Harness turns against Standard, Minimal, Anchored, or plugin candidates with frozen request-surface evidence
深海事务所 (Abyss) — an operations console for DeepSeek Harness agent fleets: every agent is a character at a desk, every line comes from the durable session log, and any past case replays and reports from disk