dsh-damage-pulse
DeepSeek Harness balance monitor with a whale-girl companion and cache-aware damage animations for every token charge.
95 results
DeepSeek Harness balance monitor with a whale-girl companion and cache-aware damage animations for every token charge.
Enhanced, persistent web search plugin for DeepSeek Harness — multi-engine routing (DeepSeek/Exa/DDG/Bing/Jina + GitHub/B站/YouTube/V2EX/小红书/Twitter/Reddit/RSS), SQLite+LRU cache, userscript-style extraction, and Playwright rendering.
Deep reading & summarization for DSH: books/papers/videos/web → structured Obsidian notes. Plugin parsers, MapReduce deep-read, JSON Schema output, idempotent cache.
Delete sessions of a DeepSeek Harness profile from the UI: header danger button + session-row menu item with a risk-consent dialog; host endpoint + tool delete the log, projection cache and workspace accounting. Works in web and the desktop client alike.
喵账单:上下文圆环弹层里实时算账——一步/一轮/会话三行各带总价,缓存命中/未命中/输出三列明细的无边框小账表,外加当前模型的平均累计花费曲线(帮你赶在花销起飞前换窗口),峰谷/一口价自动计价(价目表 rates.yml 可手填),专治『缓存命中率 100% 却越用越贵』
A kanban board plugin for DeepSeek Harness: a 'Board' tab, card comments and 15 kanban_* AI tools, with per-workspace isolation and disk persistence.
MiniMax multimodal bridge for DeepSeek Harness (DSH). One mmx_bridge tool covers describe/image/video/speech/music/cover/search/quota; optional web_search/read_image takeover; built-in client enhancement renders inline players/previews plus a settings-page management card in the Web GUI.
DSH web GUI plugin: token usage (input/output), cache hit/miss and cost statistics per conversation and in aggregate, with auto-switching DeepSeek pricing schemes
DeepSeek Harness usage dashboard with model, provider, workspace, cache, balance, and CSV insights
DSH web plugin: a usage statistics panel with per-day token trend, GitHub-style activity heatmap, cache hit-rate curve and per-model breakdown, replicating the reasonix usage stats feature.
Context compression tool (context_compact): the agent writes the replacement checkpoint itself and hands it to the host compaction engine, which skips the LLM summarizer call; automatic compaction stays on the official engine.
Enhanced ACP server for DeepSeek Harness: block-level streaming, usage/stat telemetry (cache hit rate, token speed, input/output tokens, context length, turns, tool timing), model & reasoning-effort switching, and permission-preset control over the ACP wire (Zed-friendly)
Environment housekeeper for DeepSeek Harness: toolchain inventory, agent scratch/cache scan with safe one-click cleanup, and the machine rules file (AGENTS.md) view/edit - all in the Web GUI settings. Zero runtime dependencies.
会话生命周期管理:one-shot 完成归档/删除、可续与主会话闲置归档、容量保底、连带清理 projcache——全类型生命周期管理。
DSH web plugin: show every session's theme in the left sidebar on open — warms the session projection cache at startup so session.list carries titles for cold sessions.
Codex-style context visibility for DeepSeek Harness: live context-window occupancy, composition breakdown, KV-cache hit rate, and a GitHub-style 53-week usage heatmap
Cross-session usage & cost observatory for dsh — daily token charts, cache hit ring, 90d/1y heatmaps, model and project breakdowns, adaptive color themes and CSV export in the web UI
Context Assembler DSH V0.99 — Context Assembler plugin for DeepSeek Harness (dsh): context compaction, cache-friendly topic-block management, water-pressure topic splitting, tool trace/rewrite, handoff planning and reality recall injection.
In-place three-decimal cache-hit percentage plus a widened composer stats line for DSH Web.
Accurate cache-hit rate, token usage, and cost estimates for the DeepSeek Harness Web UI: one-decimal hit rate, bucket breakdown, peak/off-peak pricing with auto price-switch, and a session usage panel.
缓存账单:上下文圆环弹层里实时算账——当前步/当前轮/会话累计三块明细,缓存命中/未命中/输出各带 token 与金额,峰谷自动计价,第三方中转照常记账,专治『缓存命中率 100% 却越用越贵』
Cost dashboard for DeepSeek Harness - model spend, tokens and cache usage across all sessions (settings section). DSH 费用看板:汇总全部会话的模型输入/输出/缓存 token 用量并按价格表计费。
DSH web plugin: pick skills from a sidebar tab — list every installed skill with global/local badges, LLM-generated descriptions (cached, never touching skill files), and inject the /skill gesture into the current session composer on check.
Per-conversation x-opencode-session header injection for OpenCode Go in DeepSeek Harness (DSH). Fixes 400 MissingSessionID, keeps prompt-cache routing optimal, scoped strictly to opencode.ai, runtime-switchable without restart.