产品成本分析：用 Agent Skill 汇总总成本与构成占比
SkillHub 社区 Skill「产品成本分析」（@user_d82fedf3/product-cost-analyzer，v1.0.1）面向实体产品与报价场景，在无需 API Key 的前提下，帮助 Agent 计算总成本、单位成本，并拆解原料、包装、制造费用、管理费用、运费、税金等构成及占比。文章说明技能定位、核实功能范围，并给出基于 SkillHub 官方文档的 CLI 安装命令与 Cursor 等客户端 skills 目录路径，以及依据官方简介的对话式使用步骤与适用注意。
Read Moredsh-token-stats - 本地 Token 统计与成本分析插件
DeepSeek Harness（DSH）插件新增「Token 统计」功能，通过读取本地日志实现零消耗的用量监控。核心功能包括：右下角常驻悬浮小窗，支持吸附、折叠、悬浮等多种形态及交互；设置页提供热力图、模型分布、性能时序及费用估算（含峰谷价配置）；支持按会话下钻、导出 CSV/JSON 以及套餐配额与预算告警。该插件不依赖外部服务，需 DSH 版本 >= 0.1.7-alpha.1 及 Node.js >= 22.15。
Read Moredsh-token-usage: Track Historical Token Usage of Local DSH Sessions
`dsh-token-usage` is a local plugin for DeepSeek Harness (DSH) designed to address the difficulty of tracking global Token usage. It features cross-session aggregation, global deduplication by ID, multi-dimensional grouping, and custom date range filtering. Data can be retrieved via the settings page, a local HTTP API, or command-line reports. It only counts assistant messages, has no extra dependencies, and helps developers accurately track costs.
Read MoreDeepSeek Harness Plugin: OpenRouter Provider Routing and Cost Analysis
xuan666-lab/dsh-openrouter-provider-advisor is a DeepSeek Harness (DSH) plugin designed to automatically evaluate and switch OpenRouter providers. It ranks providers by comprehensively considering cost, speed, context, and reliability, and offers multiple strategies (such as price-first and speed-first). The plugin supports one-click switching to the optimal route, features a built-in weighted pricing model and reliability protection mechanism, and allows users to customize weights. Installation requires supported DSH and Node.js versions and can be completed via the command line. It runs with host-level permissions to ensure API key security and supports Chinese/English interfaces as well as Agent integration.
Read Moredsh-usage: DeepSeek Harness Sidebar Usage Panel
`dsh-usage` is a pluggable usage-monitoring tool for DeepSeek Harness (DSH). It centrally displays DeepSeek and OpenRouter balances, token consumption and cost curves, cache hit rates, and heatmaps in a sidebar panel. The plugin supports session filtering and CSV export, with all calculations and credentials handled locally to ensure data security. Developers can use this tool to finely manage API costs and monitor cache performance.
Read Moredsh-token-meter-mood: DeepSeek Harness Web Plugin
The DeepSeek Harness (DSH) plugin `dsh-token-meter-mood` is designed to visualize token consumption when debugging agents. It displays detailed statistics below each assistant reply, including input, cache-hit, and output volumes, using Chinese units and grouping by four digits. It shows tiered mood emojis based on the number of output tokens (e.g., from 🥳 to 😡) and supports viewing artifacts and reasoning information. The installation command is `pnpm dsh plugin --profile web add "github:firekula/dsh-token-meter-mood"`, helping developers precisely optimize costs.
Read MoreDeepSeek Harness Plugin Relationship Graph and Cost Analysis Plugin
This article introduces EaveBounty's open-source DeepSeek Harness sidebar plugin, designed to address the challenges of complex plugin dependencies and opaque resource consumption. The plugin provides three core features: a visual plugin relationship graph with node interaction and status inspection, skill catalog statistics, and cache cost analysis with time-based cost estimation. Users can quickly enable it with an installation command and directly monitor token usage, dependencies, and economic costs on the web, making it ideal for developers who require fine-grained resource management.
Read Moreelviass/dsh-cost-insights: DeepSeek Harness Local Usage and Cost Analysis Plugin
`elviass/dsh-cost-insights` is a local usage and cost analysis plugin specifically developed for the DeepSeek Harness (DSH) web interface. It aims to address the issue of users lacking a unified entry point to review Token consumption, cache hit rates, API balances, and model prices. Distributed under the MIT license as an unofficial extension, it is compatible with DSH versions `>=0.1.0-rc.7 <0.2.0` and Node.js environments `^22.19.0 || >=24.0.0`. It also supports automatic adaptation to language and theme settings. Core features include multi-dimensional statistics (by conversation, workspace, model, etc.), visual chart analysis (line charts, heatmaps), real-time price and exchange rate synchronization (USD/CNY conversion), and API balance display.
Read Moredsh-pulse: A Cross-Session Usage and Cost Dashboard for DSH
dsh-pulse is a Web UI plugin for DSH, designed to address usage monitoring and cost management challenges in multi-workspace, multi-model scenarios. The plugin provides a unified dashboard with core features including: cross-session token usage trend analysis (supporting hourly to yearly views and heatmaps), usage breakdown and ranking by project and model dimensions, cost estimation based on per-model rates (with peak/off-peak time-of-use pricing and exchange rate configuration), as well as DeepSeek official balance queries and reconciliation. Additionally, it supports session-level detail attribution, adaptive theming, and CSV data export. The plugin runs within the DSH UI plane, consuming no additional tokens and storing no secrets, and requires Node.js >= 22.15. It can be installed via the `dsh plugin` command.
Read Moredsh-whale-report: Generating Deterministic Agent Reports from Session Event Logs
dsh-whale-report (DeepTrace) is a memory-type plugin for DSH maintained by SenmuuuuW. It deterministically aggregates data from local session event logs to generate daily, weekly, monthly, yearly, and custom interval reports. The plugin covers 10 findings, including cost peak-valley pricing, token cost allocation, tool health, retry storms, dangerous operations, and key scanning. Starting from version v0.5, it includes 4 read-only IMPROVE suggestions and a VERIFY plan, generating reports with zero LLM tokens. Installation command: dsh plugin --profile web add "github:SenmuuuuW/dsh-whale-report", licensed under MIT, current version 0.5.2.
Read More