313 results
DeepSeek Harness plugin: widens the conversation column, sets the user-bubble width and composer max height, and right-aligns the session stats strip, with a settings-row UI to tune all four. Built for the DSH client plugin slot `settings.general.item`.
A native DSH Bundle that installs a user-owned, plain-language pure PTC preset for GPT-5.6 Sol, Terra, and Luna.
DSH web plugin: a floating bubble at the top of the conversation showing the last user instruction that scrolled out of view.
Detect Claude Code and Codex CLI updates, prompt the user, and run the right channel-specific upgrade
DSH plugin: Feishu (飞书) custom-bot notifications when an agent session settles (idle) or asks for input (ask_user_question). Adapted for DSH 0.1.0-rc.7 (@deepseek-ai/cordis line).
My Habits for DeepSeek Harness: edit the user-global AGENTS.md from the Web settings — injected into every session by the built-in agent-instructions loader.
DSH plugin: notify personal WeChat via ilink protocol (Weixin ClawBot) when a task completes, errors, or needs user input. Built-in QR login flow and WebUI settings panel.
Cross-platform waiting-for-user and completion sound notifications for the DeepSeek Harness web UI.
Push vague or rich, user-configurable DSH interaction states to the local Discord client as Rich Presence.
Fifth DeepSeek Harness mode with configurable Codex, Claude Code, or DSH audits, user-approved remediation turns, progressive Cordis/skill elevation, safety pauses, and Web/TUI review controls.
Publish a validated DSH plugin package (`@deepseek-ai/dsh-*` or `@lovstudio/dsh-*`) to npm, git, or tarball channels and verify it loads in the DeepSeek Harness. Use when the user asks to publish, release, or ship a plugin. 触发：发布插件 / 上架插件 / release dsh 插件。
Quick prompts dock for the dsh web GUI: a row of user-defined prompt chips above the composer — click a chip to preview and edit its text (with {{placeholder}} fields), then sync it into the input or send it directly. Prompts are stored in the official settings document, not localStorage.
Unified DeepSeek Harness connector: edit MCP servers (cordis.patch.yml) and user skills (~/.dsh/skills) from one Web UI settings page. Repository-root package: installs either via the bare git URL (the root manifest re-exports the dsh-connector/ subdir) or via github:XJungit/omdp#path:dsh-connector.
User Message Minimap: a navigation rail for long DeepSeek Harness conversations
DeepSeek Harness config sync: export/import DSH settings, API credentials, profile plugin patches and user agent presets as portable local bundles or a single password-encrypted file (AES-256-GCM + PBKDF2-SHA256).
Chat history TOC: a user-message directory as a conversation view tab with click-to-jump
JIRA open tasks panel for DeepSeek Harness (DSH): shows the current user's open/reopened issues below the composer, per-workspace project key, persistent profile bundle.
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.
DSH package manager plugin: install dsh-bundle plugins as profile dependencies and hot mount/unmount their bundle patch rows through the official watchUserPatches cordis.patch.yml reload path — no restart, no workspace/.venv second pipeline.
Self-learning habit engine for the DeepSeek Harness — detects user-correction signals, judges habits with a low-cost model on threshold, settles candidates behind a two-level human gate
Writable plugin management tab for DeepSeek Harness — toggle, enable/disable, and uninstall user-installed Cordis plugins from Settings
Edit a sent user message and resend it: fork a new branch from that message's turn and regenerate the reply with the edited text.
Interview users, then generate and export project timelines / Gantt charts (Word + Excel) in DeepSeek Harness.
LLM pre-review for sandbox-escalation approvals: an independent-context LLM gate answers sandbox escalation requests before they reach the user, falling back to the user on any failure.