dsh-cache-precision
In-place three-decimal cache-hit percentage plus a widened composer stats line for DSH Web.
561 results
In-place three-decimal cache-hit percentage plus a widened composer stats line for DSH Web.
Replay a DeepSeek Harness session at its original token cadence — an in-app playback theater with play/pause/step/speed/seek.
基元律动费用中心:账户余额与限时额度、按分类筛选的模型价格卡(含折扣价)、官方同款服务状态与平台密钥管理;凭据只存本机 host。
A DeepSeek Harness (dsh) plugin: orchestrates complex tasks across multiple models — a 'main brain' decomposes the task and dispatches subtasks to GLM / Kimi / Qwen etc. sub-agents by each model's strengths, with per-model token usage.
Per-model hourly token usage for the DeepSeek Harness: automatic recording with a Web GUI dashboard.
Session execution analytics and audit reports for DeepSeek Harness. Understand how your agent actually worked: steps, tool calls, failures, repeated actions, token usage and verification signals.
Weak-network adaptor for the DeepSeek Harness: transparent long-backoff model-stream retries (10-min cap), local response replay cache, heartbeat auto-reconnect, degraded-mode token economy, and a bilingual (zh/en) settings page. / 弱网适配插件:模型流长退避重试、本地响应缓存、心跳自动重连(默认 10 分钟封顶)、降级省 token、双语设置页。
DeepSeek Harness plugin: serve LLM calls with the OAuth tokens your local Codex CLI and Claude Code already hold, instead of a separately configured API key — plus a composer badge showing subscription usage for those routes and for GLM Coding Plan.
Provider-level web search+fetch pool for DeepSeek Harness: bypass the official LLM search channel (0 model tokens), failover/rotate across Exa/Tavily/Jina/Firecrawl/Serper/SerpApi/SearXNG/DDG/Brave with quota-aware circuit breaking, multi-key rotation, and a GUI settings page.
DSH Web 轨迹分类视图插件:「分类」标签页 = 左上角「合计」动作统计(写入/读取/删除/下载/命令/搜索/子代理/其他工具/消息/压缩)+ 思维导图画布(一次一个问题,问题画布标签切换,问题 → 方案 → 工具调用,节点可拖动、线条带惯性)。点击问题/方案收起展开子流程,点击工具查看「做了什么/输出日志/修改内容/参数/错误」。中文分类、中英文工具名匹配,保留原始「轨迹」流程列表。纯浏览器端聚合,零模型请求、零额外 token 消耗。
DeepSeek Harness (dsh) plugin for GitHub (gh), Gitea (tea) and npm CLIs: detect/install, guided auth/config (incl. npm token rotation), issue and package tools.
Out-of-tree dsh web-profile bundle: enables --host 0.0.0.0 and gates the Web UI behind a --web_token access token (cookie after first ?web_token= open). Forks @deepseek-ai/dsh-host-webserver and @deepseek-ai/dsh-web-app/startup.
输出 token 压缩:让 agent 用紧凑的穴居人语回复(省 ~65% 输出 token),代码/命令/错误保持字节精确;token 下沉点审计(caveman learn:safe fix/offload/habit/load-bearing 分类)。受 JuliusBrussee/caveman(86k★ MIT)启发。
DSH plugin: before answering a large question, show an estimate of tokens and DeepSeek API cost as an inline row in the Web chat stream, then update it with the actual usage after the answer.
Yunoseek pink skin for the DeepSeek Harness web GUI (pink --dsw-* token layer, yunoseek brand artwork, assistant avatar) plus the Yunoseek system-prompt injection: the identity prompt rendered as the first system-prompt section.
Cache-stable progressive tool discovery and dispatch for DeepSeek Harness
Persistent Token Usage insights for DeepSeek Harness, backed by a scalable durable SQLite ledger.
Token usage statistics for DeepSeek Harness: last-7/30-day per-model daily and total consumption (bar chart + donut), plus a GitHub-style contributions heatmap over the last year.
Real-time session cost readout for DeepSeek Harness: composer-dock cost preview, today usage chart, prices kept in sync with the DeepSeek official site (peak/off-peak aware), plus price_estimate/session_cost agent tools.
DSH Web 插件:基于 DeepSeek 官方 BPE 分词表,在 composer 下方实时渲染 token 状态带,显示流式 TPS、输出 token 与首字延迟,并对比上次 step 的估算与实际偏差;纯插件自包含,不改 DSH 源码
DeepSeek Harness plugin: OpenCode GO 套餐用量(三窗口额度 + DSH 按模型 token/金额统计 + 输入框迷你条)
DSH 费用统计插件:会话费用行 + 成本归因报告,llm-pricing 动态价格。 / DSH cost attribution and usage dashboard, priced by llm-pricing.
DSH plugin: sync cc-switch provider balances and track token usage & cost by day/month/total with relay-aware currency settings.
Live token cost + account balance readouts for the DeepSeek Harness Web UI