dsh-self-improved
DeepSeek Harness 长期记忆与自进化插件:L0 对话捕获 → L1 记忆提取 → L2 场景归纳 → L3 用户画像,自动召回注入 + 技能合成,纯本地。
2316 results
DeepSeek Harness 长期记忆与自进化插件:L0 对话捕获 → L1 记忆提取 → L2 场景归纳 → L3 用户画像,自动召回注入 + 技能合成,纯本地。
Use DeepSeek web (chat.deepseek.com) as a free LLM provider in DSH — no API key needed. Plugin auto-launches a local gateway that drives a persistent browser; continuous chat, tool calls, model calibration, sub-agent concurrency.
Settings UI for DeepSeek Harness (DSH) to edit each model provider's request retryPolicy — mode (normal/always), maxRetries, retryableCodes and exponential-backoff parameters — written to settings.yaml and applied by dsh-llm-retry. 模型重试策略设置:在设置面板中可视化编辑各模型提供方的 API 请求失败重试策略。
Per-conversation x-opencode-session header injection for OpenCode Go in DeepSeek Harness (DSH). Fixes 400 MissingSessionID, keeps prompt-cache routing optimal, scoped strictly to opencode.ai, runtime-switchable without restart.
Route DeepSeek Harness model-API requests through an HTTP/SOCKS proxy without modifying Harness source
dsh-ikun-pet · ikun 桌宠 —— DeepSeek Harness 永久插件:在 DSH Deep diving 回答期间,用坤宠动图填满深潜状态行下方区块,进度 0%→100%,每 20% 切换动作与文案,完成时系统级播放提示音
在 DSH web 界面自动展开工具调用卡片,支持侧边栏开关。标准可安装的 dsh 客户端插件。
DeepSeek Harness plugin: per-event task-completion and attention sounds for turn-end / approval / question / plan-review / goal-blocked / task-failure, each with its own sound (built-in synth, local audio file) and volume (Web UI)
Agent2Agent interop for the DeepSeek Harness in one package — the realtime mesh hub (project registry, immutable message history, WebSocket presence) and mesh client (per-agent connections, serial injection), the model-facing a2a_peers / a2a_message / a2a_history tools, and the /a2a command surface
Generate and display images inline in the DSH chat via API channels or local CLIs (mmx / codex / agy), and read images into structured JSON evidence (OCR / layout / semantics) on any model — with backend probing and a bundled recovery skill.
Windows-first bash tools for DeepSeek Harness: git_bash (Git Bash) and wsl_bash (WSL2) model tools with bwrap sandbox, approval mode, terminal-style UI cards, and background-job support
Unified DeepSeek Harness plugin: role-based subagent routing + per-agent evolution — prefercmd/memory as knowledge allow/deny lists, so repeated tasks start from proven commands and save tokens
DeepSeek Harness network-proxy plugin: route the agent's own network requests (web_search / web_fetch / external APIs) through a configured HTTP/HTTPS-CONNECT/SOCKS5 proxy, persisted and auto-applied at startup, with a settings UI.
ChatGPT-style message editing for DeepSeek Harness: edit a past prompt to rewind and branch the conversation, switch versions with a counter under the bubble, and browse branches in a tree.
A reproducible science workbench plugin for the DeepSeek Harness: agent-driven cells, inline figures with feedback/rerun, manifest provenance, and environment snapshots.
OpenAI Codex provider for DeepSeek Harness with device-code OAuth, Codex CLI import, token refresh, and a web settings panel.
DeepSeek Harness (DSH) bundle porting Matt Pocock's 'Engineering for Real Engineers' + 'Productivity' Claude-Code skills (SKILL.md set) into a native DSH skill plugin. Same skill bodies, DSH skill discovery.
BitFun 与 DSH ACP 交互对接 插件: expose the BitFun agent (https://github.com/GCWing/BitFun) as a subagent provider inside DeepSeek Harness over the Agent Client Protocol
把开发变成 RPG:回合/工具/todo 积累 XP、成就徽章、等级与赛季。事件流驱动,纯函数计分。
Agent Client Protocol (ACP) server plugin for the DeepSeek Harness (dsh) — drive dsh agents from Zed and any ACP client
Standalone DeepSeek Harness plugin: continual harness self-evolution. The agent persists and refines reusable prompt notes, memories, skill contracts, and subagent specs through small evidence-backed edits, with automatic refinement gates, rollback, and prompt injection.
DeepSeek Harness-native publicity toolkit: turn any long-form source into a source-grounded, platform-native promotion matrix.
An in-process trigram index that makes code search in Pi and DeepSeek Harness 20-90x faster than ripgrep, with identical results.
Codex conversation backend for DeepSeek Harness, powered by the Codex App Server with approvals and DSH tool support.