dsh-openwrite
OpenWrite 长篇小说领域后端接入 DeepSeek Harness:统一创作 Agent 预设 + 原生创作工作台
82 results
OpenWrite 长篇小说领域后端接入 DeepSeek Harness:统一创作 Agent 预设 + 原生创作工作台
Jev Decision Engine for Agent Harnesses: native DeepSeek Harness, plus OpenCode and Codex Harness through iPolloWork
Langfuse observability for DeepSeek Harness: OpenTelemetry traces, compaction, feedback Scores, and fork lineage
DSH agent preset 分发包:V4 Flash 神模式(opencode-go)——启动时自动把 router-flash preset 安装到 ~/.dsh/.agent-presets。
Miraculous Standard — unified anchored agent preset for DeepSeek V4 Pro/Flash (official API & opencode-go). Minimal-exact two-tool first-request bootstrap, model-aware Pro/Flash paths, unified context gate, epoch-aware catalog management, rc7-ready shell handling.
Deterministic context-pruning compaction backend for dsh (DeepSeek Harness) — zero-LLM summaries, /dcp command, works out of the box. Design references Opencode-DCP/opencode-dynamic-context-pruning.
AI-assisted IC design flow skills — specification, RTL, lint/CDC, simulation, synthesis, timing analysis, and signoff
DeepSeek Harness bundle: oh-my-opencode-slim style Orchestrator agent preset with model-pinned specialist subagents (explorer/oracle/librarian/designer/fixer/council).
编程陪伴 Node-host companion: opens a persistent, per-site popup window while agents work (minimizing + suspending its process tree to freeze playback while they wait), with multi-site profiles and an in-window return-to-chat button.
Free Kilo Gateway and OpenCode Zen models for DeepSeek Harness (DSH) via native dsh-llm adapters.
Validate-then-repair patterns for tool call resilience. Five deterministic repairs, repair notes, and the structural insight that tool confusion is a harness problem, not a model problem. Based on CommandCode's approach that made DeepSeek V4 Pro outperform Opus 4.7 on tool calling.
GrayPrint — live reasoning-style fingerprints for the DeepSeek Harness web chat. Compares grayscale-vs-current writing style by internal reasoning paragraphs and independently tracks Let me exploratory openers versus We need planning openers.
OpenRouter 免费模型面板:自动拉取实时免费模型清单(综合能力 1-5★ 分级、最优优先排序),一键安装进 openrouter 路由并切换当前对话模型(全局默认 + 会话级双写,推理档自动选该模型支持的最高档),免手动配置。
Zero-signup public plaza for AI agents — speak, read, discover
AWS Kiro (CodeWhisperer) adapter for the DeepSeek Harness LLM seam: Claude and open-weight models through one signed-in Kiro account
Bidirectional conversation migration between DeepSeek Harness, Codex, Claude Code, Pi, Reasonix, and OpenCode
重启 Web 服务按钮:在设置页 General 区添加一行重启按钮,宿主端用当前运行的 node + 原启动参数重启 dsh web(日志重定向、TCP 就绪轮询、启动校验),Windows / Linux / macOS 通用;不加 --no-open,交给 dsh 自己打印并打开已授权地址。
Advanced settings management plugin for DeepSeek Harness: form-level and file-level configuration editing with an open plugin registration SDK
OpenCode Go 动态模型目录插件
Free OpenCode Zen models for DeepSeek Harness (DSH): native dsh-llm adapter plugin, no API key.
跨平台个性化内容发现:五层灵魂画像(事件→偏好→觉察→洞察→灵魂)、MBTI 推断、主动破茧探索、跨平台兴趣融合、本地优先。受 whiteguo233/OpenBiliClaw(2.9k★ MIT)启发。
Load MCP servers for the DeepSeek Harness from the mcp.json of Claude Code, Kiro, OpenCode, and Codex, with user and project layers and live reload
DeepSeek Harness WSL plugin: call TypeSafe Jev / OpenRouter System One for noul, choice, and score decisions (jev_ask / jev_check / jev_rank).
QQ Official Bot adapter for DeepSeek Harness (migrated from openclaw-qqbot)