DeepSeek Harness 플러그인 — 793 페이지

Session-scoped memory tab for DeepSeek Harness, rendered beside Chat / Trajectory / Context, covering the whole /memory surface (list, query, add, rewrite, remove, merge, proposals, budgets, audit, adapters, export, import). A companion to dsh-memento: it consumes the public ctx.memory seam and registers its own routes, so it never forks upstream. Reading the tab never increments the store's recall counters. Every route call carries a deadline, so a wedged request reports a failure instead of locking the tab. Removes address entries by primary key, so two memories with identical text stay individually deletable.

DSH plugin: meme/reaction picker for DeepSeek Harness agents. No memes, no soul. Picks the right reaction image from the kagura-agent/memes library at runtime — semantic tag matching over 26 categories.

Memo plugin for DeepSeek Harness: sidebar quick capture, local SQLite storage, private GitHub repo sync, and image/screenshot support.

Memo — session memory search for DeepSeek Harness agents: cross-session full-text recall (memo_search), distilled notes (memo_remember), and corpus stats (memo_stats) over the official sessionQuery service. Zero infrastructure, local-first.

DeepSeek Harness dual-ledger cross-session memory: auto-extracted long-term facts (MEMORY.md) plus a living project progress ledger (PROGRESS.md), recalled into every new session.

Memorix memory panel for the DSH Web GUI: browse every project and memory, read full details, archive/restore and create memories — inside the session, with writes going through the official memorix CLI.

Advanced auto-memory plugin for DSH: layered memory (episodic/semantic/profile) + knowledge graph + worldline versioning + vector/RAG retrieval + incremental graph building

A persistent, model-editable memory/notes store for DeepSeek Harness. Adds memory_set / memory_get / memory_delete / memory_search tools backed by the built-in storage domain (ctx.storageDomain) so facts survive across sessions.

DSH 工作区长期记忆：会话开始注入记忆摘要、memory 工具按需读写、轮次结束自动提炼，设置面板里可查看与管理。

Durable cross-session memory for DeepSeek Harness: local SQLite FTS5 store, three model-facing tools, and a recall prompt section — no embedding service and no sidecar

DSH 专属记忆插件：RRF 四路召回 / 确认制写入 / Fernet 加密库 / 会话转录 / 维护工具（HMS 架构移植）

Cross-session project memory for DeepSeek Harness (dsh): auto-extract durable facts/decisions/preferences at session end, retrieve and inject them into future sessions. Uses the host dsh instance model route - no bundled API key.

Cross-session long-term memory for DeepSeek Harness: a schema-validated memory domain over ctx.storage plus model-facing remember/recall/index/forget tools and a memory protocol prompt section.

Session-end memory for DeepSeek Harness: distils each turn's new dialogue into durable notes, then recalls the global set plus the current workspace's in a per-agent prompt section.

Session-end memory for DeepSeek Harness: distils each turn's new dialogue into durable notes, then recalls the global set plus the current workspace's in a per-agent prompt section.

对话记忆：记忆标签页、自动命名、隐私隔离、侧栏AI名、经验自动记录。静态 profile 插件，任何预设可用，重启不消失。

DeepSeek Harness 共享持久记忆插件：所有会话共享一个记忆库，提供 memory_add / memory_search / memory_list / memory_remove 工具，并在 Web 侧边栏提供 Memory 面板

Layered memory plugin for DeepSeek Harness: strictly separated project/global memory, host-driven recall before tasks and continuous learning after them.

DSH 长期记忆插件（多 provider）：OpenAI 兼容的云端嵌入 + Qdrant 向量库，自动记录对话要点与图片概览、自动召回，提供 memory_search/save/forget 工具，并在设置页提供 provider 管理、记忆管理与定期维护。

DSH 长期记忆增强插件：文档索引 + 语义检索 + 跨会话 KV 记忆（agent 工具）

Persistent long-term memory plugin for DeepSeek Harness (dsh): single memory tool, markdown storage, auto session summary, curator governance, user-profile injection.

DSH 本体（web/headless profile）的跨会话语义记忆工具：memory_add / memory_search。智谱 embedding-3 向量 + SQLite + 余弦相似度。配置可选、绝不因缺 key / 缺库 / 缺服务而崩。

Long-term memory for the DeepSeek Harness Web GUI: project-scoped and global memory stores, a Memory view tab beside Trajectory with one-item-per-row list and an Obsidian-style link graph, automatic per-turn memory extraction, automatic memory briefing injection into every request, and small-model periodic consolidation (dedupe / merge / link / archive).

Long-term memory for DeepSeek Harness agents: user preferences, project facts, session summaries, and knowledge entries, with auto-injection at session start.