Plugins

Native LLM-provider vision bridge: images pasted in the chat are described by a vision model (Qwen3-VL via pi-ai/llama.cpp) and the text description is fed to text-only DeepSeek for the reply — image admission, routing and compaction all run through harness-native mechanisms, with an LRU description cache and 503 retry.

Native DeepSeek Harness plugin: feeds every prompt into the local agent-feedback-loop capture pipeline and injects compiled recurrence rules back as context.

OpenBKN business knowledge-network context for DeepSeek Harness

DSV4.1 reasoning support with multimodal context and bounded artifact acceptance for DeepSeek Harness

IDE-style Git panel in the DSH Web GUI: a resident tab with a three-column overview (branch/tag list, searchable commit-history graph, commit detail with changed-file tree and message) and a changes page (stats bar, uncommitted list with per-file stage/unstage/discard, commit box with amend); side-by-side diffs carry lazy highlight.js syntax highlighting, on-demand between-hunk context expansion and word-level intra-line emphasis, with image files shown as before/after panes; plus a zsh-style branch marker in the input bar. Every git command runs as an argv array through the host subprocess service, cwd locked to the repo root.

Select text in the DSH web UI to stream a translation and an explanation of what it means in that specific context, read against the conversation it came from; the panel keeps a replayable history and can be promoted into a full session.

Web reading tools for text-only agents. Five read-only tools: web_fetch (any URL rendered as clean Markdown or plain text), web_links (inventory of a page's links), web_feed (RSS 2.0 / Atom entries), web_headers (HTTP status, response headers and redirect chain without downloading the body, HEAD with automatic GET fallback), and web_table (HTML tables extracted as structured rows with header detection and colspan/rowspan grid expansion). Transport is Node fetch plus a zero-dependency HTTP proxy implementation (CONNECT tunnelling, absolute-URI requests, NO_PROXY with suffix/wildcard/IPv4-CIDR matching, Proxy-Authorization). Zero runtime dependencies.

Universal Model Context Protocol (MCP) bridge for DeepSeek Harness: connects agents to stdio/sse/websocket MCP servers with a 10k dormant scale engine, sub-millisecond BM25 tool search, dynamic token budgeting, and a real-time web UI dashboard.

Comet-mode animated caret for the DSH Web composer: a glowing trail glides to the text position, with configurable accent color, thickness, trail, and a 500 ms idle blink. Extends the surface to ask-question card inputs, and fixes the caret vanishing on an empty composer or first focus, forward drag-selection not following the pointer, soft line breaks measuring against the previous line, and the caret painting outside the composer when long input scrolls. Maintained fork of Lacquervii/smooth-cursor.

DeepSeek Harness (DSH) 推理等级自动选择插件：由 Jev System One 模型判断每条消息值多少思考量，按模型声明的等级自动推导档位，上下文信封让「继续」这类追问继承话题深度，低置信度向上取，任何失败都静默沿用原等级。 | Jev-driven reasoning effort per message: per-model ladders derived from what each model advertises, a fixed-size context envelope so follow-ups inherit topic depth, ties break upward, silent fallback on every failure path.

Vision bridge for text-only DeepSeek routes that analyzes attached and local images through configurable OpenAI Responses, Chat Completions, or Anthropic Messages endpoints while leaving image-capable routes native.

Windows minimal agent preset for DeepSeek Harness: fixed one-line persona, three tools, no runtime context.

Per-message delete for DSH: remove a user message, one reply step, or a whole assistant reply from the derived model context through the official surface-replace contract, and hide it from the visible transcript without rewriting the append-only log.

Host-level proactive wake-ups for DeepSeek Harness: the model schedules its own alarms (once, recurring interval, or cron, with optional random jitter), cold sessions are woken on time by session resume, model-initiated follow-ups are gated by quiet hours and a daily delivery budget, silent wake turns end with proactive_reclaim and fold out of model context as tombstones, declared schedule files glob-sync JSON schedules into host alarms idempotently, min_idle_seconds defers wakes until the destination session quiets down, and a web panel (settings section plus per-session tab, SSE live refresh) manages alarms and wake history — the heartbeat engine for AI companions and roleplay characters with their own daily routines and world evolution.

Merges other sessions' context into the current turn: evidence-weighted conflict resolution, retraction of claims whose premises were retracted, trust labelling that resists cross-session prompt injection, and budget-bounded packing. Ships a benchmark.

Feeds every dsh prompt into the local agent-feedback-loop feedback-memory pipeline — captures dissatisfaction, reviews it offline into reflections, and attaches the compiled recurrence-rule context back to the turn, never blocking a prompt. Requires the agent-feedback-loop runtime; running agent-feedback-loop install afterwards wires existing dsh profiles automatically.

Shows a DSH session as a conversation canvas — every turn is a node, any turn forks a real child session, and a compacted context folds into one green node you can keep talking from; the left rail groups canvases by DSH workspace and renames them in step with the session titles.

Same-origin in mechanism with the two loudest lines in context compression. The code-skeleton gate follows the skeletonization approach of Headroom (Apache-2.0), whose published headline is 20% fewer tokens for coding agents and 60–95% fewer tokens for JSON, same answers. The estimator channel follows TokenPilot (arXiv:2606.17016), which reports up to 60% lower cost for long-session agents. Both figures are theirs, quoted as-is; this plugin ships no benchmark of its own and claims no reduction of its own. What it adds for DeepSeek Harness: choose a compression profile, set the Auto Compact trigger level and toggle code-skeleton compression from one settings section, with exact DeepSeek V4 tokenizer measurement, same-revision count verification, and fail-open behaviour that keeps the original tool results on unsupported models.

Team knowledge base for offline knowledge sharing (FDE box scenario): browse, full-text search and refinement entry points; raw material is queued automatically and distilled into wiki pages by a serial bot session (Karpathy LLM Wiki pattern: immutable raw, two-step refinement, log stream, monthly lint).

Session instruction anchor for DeepSeek Harness: the first step of a new session injects a freely composed instruction block as a plugin-sourced message, and it is re-anchored after a summarizing compaction, every N turns, or when metered context pressure crosses a threshold. An /anchor command re-anchors on demand without a model call; a settings page manages the paged preset library, injection order, re-anchor source and character limits.

SiYuan (思源笔记) notes integration: 17 siyuan_* host tools covering notebook and document listing, full-text search, SQL query, reading documents and blocks, creating and appending content, moving and renaming documents, daily notes and block attributes, plus a Web settings page for the SiYuan URL, API token and per-tool switches.

Multi-language codebase intelligence for MCP-compatible IDEs, the CLI and DeepSeek Harness: chat, search, audit, refactor, and board-ready reports from local code.

One-click Concise output style for DeepSeek Harness: results first, no filler, same work depth. Every final reply opens with a plain-language digest card (select text to copy), scoped per session, working in headless/CLI too, with /concise command, style.md overrides and a local API. Links and file paths inside the digest card render as clickable chips: URLs jump, local files open with the OS default app (v0.9+). A per-step pre-generation contract reminder (v0.11.2+) measurably fixes digest misses on audit-report/table-shaped replies.

Per-workspace MCP scoping for the DeepSeek Harness — decide which MCP servers a session may see from the directory it was opened in.