메모리 및 지식 · DeepSeek Harness 플러그인 — 3 페이지

Shell plugin: three-column AppFrame with drag handles, ctx.layout viewing-state service (navigation + panels)

The Web feedback surface: per-message Like/Dislike in the assistant-message action strip and the feedback dialog behind both ratings and /feedback, backed by the messageFeedback and sessionFeedback Host Remotes

Model selection over the shared model catalog, Session projection, and session.selectModel

Web Session-header "Open In..." split button opening the session workspace directory in a locally installed application

Permission surfaces: a new-session default in General settings and a current-session /permission popup over the permissions projection

Plan-mode composer control: the conversation.input.plan seat over the plan projection and the /plan command channel

Pure React atoms for the dsh web UI: controls, icons, markdown, and JSON inspectors (zero cordis)

Unified Web @file and @session reference source

Browser UI renderer: React slot bindings, ctx.uiRenderer, and the assembled application root

Read-only active Schedule catalog in the Web Session header

Session Controller adapter for React and session-scoped slots

Immutable command-line handoff from a dsh launcher to any app plugin that injects cmdlineArgs

The dsh one-shot bundle: a direct core Agent/Session runner over dsh-base with no Host, HTTP, or browser layer

The dsh SDK profile bundle: stdio JSON-RPC serving and process lifecycle over dsh-base

The standalone minimal SDK profile bundle: JSON-RPC, one DeepSeek adapter, persistent shell, and JSONL sessions

The dsh browser-surface bundle: the web patch layer over dsh-base plus the runtime glue plugin (frontend dist serving, web-surface prompt, bash runtime variables, URL line)

Three-tier memory control plane for DeepSeek Harness: persistent runtime context, searchable project documents, pluggable long-term memory, smart routing, supervised agent workflows, WebUI, and headless tools.

Curated file-based long-term memory for DeepSeek Harness — discovery manifest for the installable plugin in adapters/dsh/plugin/

为 DeepSeek Harness 带来分层记忆（全局 / 用户 / 项目 / GIT 分支 / 每日）与自我进化（经验沉淀 + 技能自动创建）和技能管理、待办管理、CLI 调度（kimi/codex/grok/hermes 等外部 AI 统一调度）、临时信息便签，带 WebUI 管理界面。Hermes-style long-term memory, self-evolution, skills, todos and external CLI agent dispatch for DeepSeek Harness.

Tavern mode for dsh web — SillyTavern character cards, prompt presets, lorebooks, personas and regex on the dsh agent runtime.

Structured memory engine for DeepSeek Harness. Offline semantic search, entity-attribute-timeline, autoDream self-consolidation, and human-editable Markdown storage.

Bounded, layered, approval-gated, auditable cross-session memory for DeepSeek Harness — a capability seam (ctx.memory service + local SQLite provider + memory tool + frozen snapshot injection), not another memory warehouse

Local image-code knowledge library with native macOS and local Web clients, plus MCP and Skill for Pi, DeepSeek Harness (dsh), Claude, Codex, Cursor, and Wisp.

DeepSeek Harness plugin that wires DSH agents to OpenContext for durable memory + retrieval-augmented context.