Plugins

Ghost-text completion for the composer: FIM suggestions stream in inline after a typing pause (SSE, first token rendered immediately), Tab accepts one word at a time and Shift+Tab takes all, in-flight requests are truly cancelled on retyping or Esc. Plus two-layer typo checking: a browser-local dictionary scan (Chinese wrong phrases, English misspellings and proper-noun capitalization, user-defined per-browser entries) with a debounced LLM context pass, underlined in red in the text with per-item navigation and click-to-fix.

Shared Markdown memory for DSH and other coding agents, with automatic capture, pre-step context injection, searchable recall, and memory-to-skill self-evolution through a review panel.

Per-round cache billing inside the composer context-meter popover: what the current call spends on cache hits, misses, and output in CNY, with automatic official peak/off-peak and per-model pricing; hidden on non-official DeepSeek routes.

A deterministic block-world skin and context castle for the DeepSeek Harness web GUI.

High-fidelity, faithful, bilingual, recursive compaction backend for DeepSeek Harness — a drop-in upgrade over dsh-compaction-basic.

Reads the session files thirty-three other coding agents on this machine already wrote — Claude Code, Codex, Cursor, VS Code Copilot Chat, opencode, OpenClaw, Hermes, Kimi, Cline, Zed and more — including sessions from before it was installed: six tools (deja_recall, deja_session, deja_blame, deja_fix, deja_how, deja_remember), a /deja command, and optional automatic recall added to the runtime context. Local BM25 index, no LLM, no embeddings, no network (dsh plugin --profile web add dsh-deja).

Token attention management panel for DeepSeek Harness: per-task/day/week/month token usage and cost tracking (input hit/miss, output, reasoning), DeepSeek peak/off-peak pricing, task-type recognition, and advice on when to switch sessions, write a hand-off, or compact context.

DeepSeek Harness 的 MCP 服务管理器：在 设置 → 插件 → MCP 标签页中发现 GitHub 上的 MCP 服务仓库（topic:mcp-server / modelcontextprotocol），探测 npm / PyPI 发布状态与 README 启动线索，详情查看后一键安装 / 移除。服务以 @deepseek-ai/dsh-mcp-client 行登记进 cordis.patch.yml，HMR 热重载，无需重启。支持 GitHub 代理。

Conversation and workspace undo: use the /undo timeline, session-header action, or message icon to roll back before any sent prompt; private Shadow Git snapshots restore files, the reverted turns stay out of model context, and Undo Rollback can recover mistakes. Includes archived-session management and per-tool colored cards.

Workspace/chat context menu for the DSH home page: pin, rename, open in file explorer, archive, fork, copy, and open in a new window.

Browser Web Speech dictation for the Composer: recognition needs no dedicated ASR server, key, or model download; reuses Session text and a configured DSH model for contextual phrase hints and optional transcript polishing.

领域建模：主动构建并打磨项目的领域模型（CONTEXT.md 术语表 + ADR 决策记录），挑战模糊术语、推敲边界场景。受 mattpocock/skills（223k★ MIT）启发。

Local-first usage and cost meter for the DSH web UI: composer button with live session cost and balance; panel with a stacked per-model daily/7-day chart priced per effective list-price interval (official pricing page synced daily), monthly usage tiers with a canvas share card, 30/50/85% context-pressure warnings with one-click /compact, budget alerts, and an in-panel npm update check with one-click update.

上下文工程：按层级（规则文件→规格→源码→错误输出→对话历史）主动策划 agent 看到的信息，提升输出质量。受 addyosmani/agent-skills（88k★ MIT）启发。

Content-aware tool-result reducer: 13 reducers cut oversized output by meaning at tools/post-execute (failing tests kept, diff context dropped, JSON and CSV sampled, lint problems grouped by rule, dependency stack frames collapsed, crowded directories counted), with full originals spilled to disk behind a locator.

Cross-session memory plugin for DSH (v0.6.1, npm @asherliner/dsh-memory-connect): auto-extraction, SQLite FTS5 indexing, RRF semantic recall injected into the system prompt every turn, scheduled maintenance, LLM-powered consolidation via ctx.llm, context explosion prevention, and global Soul identity via ~/.dsh/soul.md. v0.5.0 added temporal context graph (valid_from/valid_until/supersedes, append-only corrections via reviseMemory), trust model (recalled history injected as untrusted reference, current instruction always wins), turn-end summarization. v0.6.0 added local semantic embeddings (BAAI/bge-small-zh-v1.5 via embed_server.py, cosine + RRF fusion with FTS5, no cloud calls, ~232 msgs/sec on CPU). v0.6.1 fixed ~ expansion in path config. Zero-config install.

Project memory for dsh with native task-system integration: session todos and file reads auto-condense into cross-session task records (task↔file links), so workflows are switchable and resumable without re-contextualizing. Read-time memory (no upfront scan): extracts code symbols, summarizes docs, and cross-links docs↔symbols; optional TS semantic enhancement (auto on when TS installed). Experience notes auto-dedup. Every recall includes a traceable citation. Pure JS, single dep, works out of the box, silent operation, zero overhead.

Codex-style side chat & selection annotations for DeepSeek Harness (DSH) web — fork the session into a persistent side panel; quote selections into context. Thin consumer of dsh-better-sidebar.

Native DeepSeek Harness plugin for IsingQ guided QUBO modeling and solving

Practical patches for missing DeepSeek Harness Web context and history controls.

Cross-session recall for dsh web: '@' mentions and a composer picker link previous sessions as reference chips; the host snapshots them in as recall context right before the question, recorded read-only in a session-header chain badge

Auto-compact a DSH conversation after the current task finishes, at a user-set context threshold. Settings-page switch + slider included.

Local-first code knowledge graph for coding agents: 10 MCP tools for layered context compression, DAG planning, incremental AST indexing, a cross-session skill flywheel, diagnostics, and graph artifact import/export.

MCP client hub (stdio / Streamable HTTP / SSE) with a web console for DeepSeek Harness — manage MCP servers, agent skills, slash commands, project & global memory, and prompt sections.