@weibaohui/context-razor
dsh 插件 · 上下文剃刀:把当前会话上下文逐条列出(角色/预览/≈token,cl100k 估算),超阈值标红,勾选后精确裁剪——不经 LLM 总结,删了什么一目了然。
856 results
dsh 插件 · 上下文剃刀:把当前会话上下文逐条列出(角色/预览/≈token,cl100k 估算),超阈值标红,勾选后精确裁剪——不经 LLM 总结,删了什么一目了然。
What can this MCP server actually do? Inspect its real capability surface, seal it, and get told when it changes. Never invokes a tool; gives the server a minimal environment.
Manual budget-targeted context compression for DeepSeek Harness: the conversation model picks ranges, a cheap flash-tier route writes the checkpoint summaries.
Eyes for text-only DeepSeek: view_image tool (any OpenAI-compatible VLM, local Ollama or cloud) + chat image-attachment bridge that rewrites pasted/dropped images into view_image path markers
Keeps long coding-agent sessions lean: wrap work in named tasks and, when one is done, fold its whole span into a short titled summary. The conversation stays readable, context costs stay low, and every fold's original content can be read back on demand. For DeepSeek Harness.
DeepSeek Harness bundle that registers the session-summarize skill pack on ctx.skills.
DeepSeek Harness (DSH) native plugin: the describe_image tool, a vision bridge (image -> mimo-v2.5 -> text description) over the ctx.fs / ctx.credentials seams
Qwen-MM-Plugins integration bundle for DeepSeek Harness: bundled multimodal skills, MCP server rows, and an image attachment bridge for text-only model routes
元压缩:模型自己决定何时、如何压缩自己的上下文——列 surface、选区间、以自写文本替换;替换走官方 compaction 事务(检查点、配对平衡、可重建)。Meta-compaction: the model directs its own context compaction through the official compaction seam.
Unofficial bounded peer discovery, messaging, receipts, and safe context projection for DeepSeek Harness agents.
Quote selected assistant text into your next request: a Host-side annotation service (session events + Typert Remote) plus a Web client plugin (quote button + pending-quote dock).
Out-of-tree dsh plugin: '@' project file/directory suggestions and plain-text references in the chat composer.
MCP 服务器管理器:在 DSH 设置页配置 MCP 服务器,宿主侧按配置热挂载/卸载 mcp-client 实例
A DeepSeek Harness (DSH) plugin that reads PDF files of any size (no 64KB limit): full Unicode text extraction (Chinese/English) via pdfjs-dist, with automatic OCR for scanned or image-heavy pages (Windows WinRT OCR zh-Hans-CN + en-US, or optional tesseract.js).
Per-directory context composition for DeepSeek Harness — .dsh/context.yaml declares systemPrompt + preload for every agent loop; memory is a baseline, not a preset
Multi-provider image generation and editing tools for DeepSeek Harness
Cairn — long-term memory for the DeepSeek Harness: persistent cross-session memory (store, tools, auto-extraction, context injection) as one installable profile bundle
dsh-eye-vision: give text-only DeepSeek Harness models eyes — image understanding via any OpenAI-compatible multimodal API. Fork of dsh-free-vision with custom-provider fixes and an allowed-directories whitelist.
Fold each DSH Web turn's thinking, tool calls and context into compact expandable process boxes cut at every piece of model text, and reuse the shipped turn-process disclosure row as the toggle whenever it exists.
AiToEarn content-creation suite as a DeepSeek Harness plugin: creative director, script writer, image-text and video generation, and Douyin publishing.
Static and runtime security guard for the DeepSeek Harness: scans plugins and workspaces for malicious code, context injection and token waste (block/warn/clean), intercepts dangerous runtime tool calls and prompt steps, and exposes /scan, plugin_scan, the scan web panel and a user-managed allowlist. Fully static analysis — never executes scanned code.
便签 (sticky notes) client plugin for the dsh webui: floating, resizable post-it notes with markdown text and image notes, per-conversation binding, archived in browser localStorage.
Web sidebar indicator: API key context headroom, account balance, and estimated turns until compaction
Speech plugin for DeepSeek Harness: per-message speak button, composer voice input, and auto-announce toggle, over cloud TTS/ASR with Web Speech API fallback