dsh-plugin-use-images
Use image-generation HTTP APIs as native image models in DeepSeek Harness
171 results
Use image-generation HTTP APIs as native image models in DeepSeek Harness
Grant official-style reasoning-effort selection to hand-declared pi-ai models (DeepSeek relays and the like) in the chat box
DeepSeek Harness profile bundle that adds Grok 4.6 to the xAI model catalog.
Global and workspace rules for DeepSeek Harness agents: configurable storage paths, system-prompt injection and a rules_read model tool.
在 dsh 模型目录的每一行直接设置该模型的思考强度范围与支持的输入模态 (Per-model reasoning-effort range and accepted input modalities, right in the dsh Models catalog)
基于Model Context Protocol (MCP)协议的全网热点趋势一站式聚合服务,支持Python实现,适用于新闻资讯、社交媒体、科技开发等多领域。
一个提供实时《Old School RuneScape》玩家统计数据和排行榜数据的Model Context Protocol (MCP)服务器,支持多种游戏模式和玩家比较功能。
数学建模竞赛 Agent(DeepSeek Harness preset)v7:CUMCM 国赛专项,资产接管→Discovery+竞争搜索→方法学契约→代码图表→编辑/视觉证据链→论文→全门禁验收,含 CUMCM Typst/LaTeX 双模板与 2026 国赛规则 authority
DeepSeek Harness plugin bundle: the 墨墨学习 mirror MCP server (mcp__maimemo__*) plus the momo-daily-learning skill
DSH plugin: when an upstream model API request fails, dump the exact raw request body and the raw upstream error response to disk.
Type-to-search filtering for DeepSeek Harness model selection lists
DeepSeek Harness plugin: LLM model selection & deployment analysis (deployability, VRAM, TTFT, latency, throughput, power for 38 models × 20 GPUs/NPUs)
Evidence-backed business strategy, pricing, profitability, elevator pitches and business plans.
DSH web plugin: one live model and timing summary during every output
补齐 DSH 的 OpenCode Go 模型目录
一个基于OpenRPC的Model Context Protocol (MCP)服务器,提供JSON-RPC功能调用和方法发现服务。
让主会话模型从已配置模型里自主选择子代理模型
EN: Compatibility and safety adaptation layer for non-DeepSeek models in DeepSeek Harness Web. ZH: 面向 DeepSeek Harness Web 非 DeepSeek 模型的兼容与安全适配层。
dsh addon: a Local Models settings tab that starts and stops a llama-server child process living with the dsh host process.
Nushell-only shell executor, dialect preflight, and Nushell syntax teaching for DeepSeek Harness (dsh 0.1.7-rc.1): every ctx.shell execution runs as `nu --no-config-file -c`, other shells are refused, bash/PowerShell habits are refused before they spawn (including the ones Nushell silently misreads, like `2>$null`), a failed call carries an actionable Nushell hint verified against captured real Nushell stderr, and the model is taught Nushell.
Model availability tester panel in the Models settings page: one-click real streaming test per added model with TPS / first-token / elapsed metrics
Expose per-model thinking levels (reasoningEfforts) inside the Models settings card for a custom model API route.
Route selected models through an outbound proxy in DeepSeek Harness, falling back to a direct connection while that proxy is unavailable.
A DeepSeek Harness plugin that registers a greet tool callable by the model.