dsh-web-artifacts
SQL 优化模式:EXPLAIN 分析、索引策略、N+1 解决、查询改写。受 wshobson/agents(38k★ MIT)启发。
135 results
SQL 优化模式:EXPLAIN 分析、索引策略、N+1 解决、查询改写。受 wshobson/agents(38k★ MIT)启发。
SQL 优化模式:EXPLAIN 分析、索引策略、N+1 解决、查询改写。受 wshobson/agents(38k★ MIT)启发。
把技能当可训练参数:像训练神经网络一样训练 agent 技能(epochs/batchsize/learning rate/验证门禁,但不碰模型权重)——rollout→reflect→aggregate→select→update→evaluate 循环、候选编辑仅在严格改善 held-out 验证分时接受、文本学习率预算、零推理时模型调用、部署紧凑 best_skill.md。受 microsoft/SkillOpt(MIT)启发。
Repository and agent-runtime analysis skills as an installable DeepSeek Harness plugin: repo recon, DSH enhancement analysis, runtime optimization, publish screening, and DSH plugin-potential assessment.
DeepSeek Harness web UI plugin: tool-call value graph over conversation trajectories — call cost/count/duration weights, bottleneck split, and next-turn optimization prompts.
把长文本渲染成图片发送来压缩 LLM 输入 token(DeepSeek 视觉模型、每图 384 token 封顶)— text-as-image token compression for DeepSeek Harness
Mobile/touch optimizations for the DSH Web GUI: iOS input-zoom fix, safe-area insets, touch targets, and viewport clamps for stock dialogs on narrow screens.
Mobile & iPad touch optimization plugin for DeepSeek Harness Web GUI with WebKit fast-tap bypass, hover delay elimination, keyboard focus guard, and smooth workspace scrolling
DSH plugin for token usage analytics, cost tracking, and optimization signals
DeepSeek Harness plugin: aggressive token optimization without affecting task quality. Trims tool results, manages context window pressure, optional LLM-summary compaction, stable prompt-cache layout.
DSH 对抗多智能体三层失效的防御插件:读取台账去重度量(read-ledger)、免疫压缩规则区+完成机械门禁(immutable-core+completion-gate)、低成本审查lane(role-router)。零外部依赖,注入即用。
DSH Prompt Optimization plugin: one-click prompt optimization in the composer input, with an optional dedicated optimizer model in Settings
DSH Effort Router(模型分流):按每一轮请求的难度自动选择模型与思考强度——简单问题走便宜快模型,难题才叫强模型。规则判定零 token,灰区交给一次小模型裁定,图片档强度随难度升降(low/high/max)。每轮请求级覆盖,不改你的会话选择与默认模型。
DSH 提示词锻炉:零配置流式优化(复用会话模型)+ 轻量词库(JSON 文件)+ 一键存技能。Prompt Forge: zero-config streaming optimization + file-based prompt library + one-click skills.
Multi-Agent collaborative skill forging system for DeepSeek Harness — distill conversational experience into verifiable, reusable skills with evolutionary optimization.