49 results
On-demand vision for text-only LLMs in DeepSeek Harness: keep pasted/attached images in the backend, and let the model call view_image to look at them anytime via any OpenAI-compatible vision endpoint — as if the model were natively multimodal.
Set the default initial directory of dsh's workspace directory picker (native + browse backends), with a GUI settings card. 设置 dsh「添加工作区」目录选择器的默认初始目录（含 GUI 设置卡片）。
High-fidelity, faithful, bilingual, recursive compaction backend for DeepSeek Harness — a drop-in upgrade over dsh-compaction-basic.
Riff winning short videos — give one source (a TikTok link, an uploaded video, or an analyzed template) and the backend riffs its emotion formula into your own AI video (post-ready short-form or UGC-style ad creative), with optional digital character, product placement, and language.
An independent todo board for DeepSeek Harness: list/board/calendar/week/project views, an AI tool (todolist), and a self-contained storage backend. Registers a top-level Todos tab; registers a right-sidebar Today tab when dsh-better-sidebar is present, otherwise provides a standalone Today sidebar.
Text-to-image for DeepSeek Harness: a `draw_image` model tool, an input-bar 生图 button with a prompt popup (async generation, 4-grid results, download / keep / delete), an /imgdraw image route, and persisted history. Backends: DashScope wan2.7-image (free default) and SiliconFlow Qwen-Image.
Active Context Pruning (ACP) for the DeepSeek Harness — model-driven context management as a CompactionEngine backend.
HashiCorp Vault backend for the DeepSeek Harness credential seam: central secrets, AppRole machine auth, rotation without restart, and no long-lived provider key on the agent host
Memory Tree bridge for DeepSeek Harness — search/persist memory, manage the memory tree, and drive the local/cloud extraction backends. · 记忆树桥接：检索/落盘记忆、管理记忆树、本地/云端提取开关。
Git worktree isolation backend for DeepSeek Harness workflows: binds each write-capable workflow agent to its own recoverable checkout.
Container-isolated backend for the DeepSeek Harness code-execution seam: Code Mode programs run in a fresh container with no network, a read-only rootfs, and kernel-enforced memory, CPU, and pid ceilings
AgentFrame compaction backend for DeepSeek Harness: semantic + physical KV compression (28.4x) replacing LLM summarization
Multi-backend web_search provider for the DeepSeek Harness web seam (ctx.web): Exa (REST with key, anonymous hosted MCP without) and Firecrawl (v2 search API) today, extensible to more backends (SearXNG, ...), with automatic failover when one backend rate-limits
Mobile remote control for the DSH web GUI with TunnelMux as the tunnel backend: scan-to-pair QR entry, one-time pairing tokens, live device status, revocable mobile sessions, and an auto-tunnel driven by the TunnelMux control API (127.0.0.1:4765) instead of an embedded cloudflared binary.
DeepSeek Harness plugin: run a live security health check on the deployment (credential file permissions, process-env secrets, permission/approval mode, telemetry, anonymous id, LAN exposure, sandbox backend) and get an actionable report mapping each finding to upstream discussions.
为 DeepSeek Harness Web 页面左下角添加“重启后端”按钮，一键重启 dsh 后端进程。
High-fidelity, faithful, bilingual, recursive compaction backend for DeepSeek Harness — a drop-in upgrade over dsh-compaction-basic.
Paste images into DeepSeek Harness chat and have them read by a free local backend (macOS Vision / Tesseract) before the text-only DeepSeek model answers
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。
MySQL-backed storage backend for the DeepSeek Harness storage hub (ctx.storage)
Local OCR for DSH: handwritten Chinese and math formulas to Markdown with LaTeX. GPU (DirectML) / CPU / NPU backends, settings-driven, one-click install.
Docker-container (WSL2 backend) shell executor for the DeepSeek Harness bash capability seam on Windows: every bash tool call runs in a disposable Linux container with the workspace bind-mounted.
Fork-style Git GUI packaged as a DeepSeek Harness plugin (Express backend + browser frontend).
Tavily-backed web_search / web_fetch providers for DeepSeek Harness — replaces the built-in DeepSeek search backend.