334 results
Session topic titles for DeepSeek Harness: LLM-generated titles for sessions stuck on the prompt-excerpt fallback, retry on every turn end, plus a /topics-size endpoint for the hover-card size patch (out-of-tree plugin)
DeepSeek Harness plugin: fingerprint the provider, model id, sampling, offered tool schemas, and system-prompt prefix of each request, and diff the live card against a pinned one through the modelprint_check tool
DSH 工具集成插件集：为 Agent 提供 Prompt 结构化优化等实用工具。
DeepSeek Harness dual-face plugin: a sparkles button beside the model selector that polishes the current draft with the selected model, plus a settings section for style (concise/detailed/formal/casual) and detail (3/5/expand).
DSH plugin that forces the agent's chain-of-thought and reasoning to stay in Chinese via a per-step system-prompt injection — 让 Agent 的思考过程也自动使用中文
DeepSeek Harness (dsh) plugin: manage named roles, each with a preset system prompt. The active role is injected into the system prompt (replacing it before the conversation starts, appending afterwards). Ships a host service plus a Web client UI.
开口即成文 · Speak-to-prompt for DeepSeek Harness：云端 ASR 语音识别 + 提示词优化 + 填入草稿/自动发送，跨平台 macOS / Windows。
DeepSeek Harness web UI plugin: tool-call value graph over conversation trajectories — call cost/count/duration weights, bottleneck split, and next-turn optimization prompts.
DeepSeek Harness plugin: make the chain of thought open with "We need" instead of "Let me"
High-fidelity, faithful, bilingual, recursive compaction backend for DeepSeek Harness — a drop-in upgrade over dsh-compaction-basic.
轨迹视图插件：在 DeepSeek Harness Web GUI 的「轨迹」标签旁新增可回放的横向时间轴——minimap 总览、搜索、回放、轮次跳转、跟随模式、历史分页、步骤/子调用耗时与吞吐、提示词变更对比、原始请求分析与 Markdown 导出。
Declutter the DeepSeek Harness chat: hide agent execution-log rows (Think / context injection / tool calls), tighten message spacing and font sizes, wrap messages in rounded-corner bubbles, and inject an editable "reply concisely" instruction into the system prompt.
Mercury Spec Ops MCP Server 是一个基于模块化架构的动态提示生成和模板组装工具，适用于AI助手与专业内容的交互，支持31种技术栈、10种分析维度和34个模板部分的动态生成。
Press Enter to approve the sandbox permission-escalation prompt in the DeepSeek Harness WebUI instead of clicking the Allow once button.
PLUR memory for DeepSeek Harness — engrams injected straight into the prompt, no tool call required
Condition-driven session wake for DeepSeek Harness: durable file and GitHub watches that fire a configured prompt into a target session (cold-wake included), with transition-only dedupe and throttling
docgen — document workshop skill pack: pure-prompt Agent Skills for README generation, PR descriptions, changelogs and code review, installable into dsh (DeepSeek Harness).
Model-facing describe_image + extract_text tools for the DeepSeek Harness web GUI: gives a text-only model reliable image understanding and OCR through an OpenAI-compatible vision endpoint, with critical-inspection prompts, auto-preprocessing, retries, and a persistent answer cache.
Image auto-describe provider for the DeepSeek Harness apiproxy admission seam: transcribes pasted images through configurable vision routes (Qwen first, GLM fallback) so text-only session models still receive image prompts.
Turn watchdog for DSH: detects live sessions whose turn looks stuck (no turn/end within a threshold), injects a quiet warning into the prompt context and exposes a status tool. Zero runtime deps, no custom events.
DeepSeek Harness plugin: injects an [IMPORTANT] chain-of-thought rule block at the very top of the system prompt of every model request (English CoT starting with I'm, no 'Let me', direct verbs).
SillyTavern feature port for DeepSeek Harness: character cards, world info (lorebook), and generation presets as prompt sections, runtime context, request overrides, and slash commands
在每个 DSH 新会话的 system prompt 开头自动注入可自定义的固定提示词，并提供输入栏 UI 编辑入口
Mixture of Agents (MoA) tool plugin for DeepSeek Harness — run several models in parallel on the same prompt, then have a stronger aggregator model synthesize a final answer. On-demand tool, zero cost when unused.