Memory

Adds a knowledge-base sidebar backed by the separately installed local Obsidian Quote service: select registered vaults or folders, retrieve read-only note excerpts and cited context, extract rules, generate project convention briefs, and view shared citation statistics.

Zhinao (second brain) for DeepSeek Harness: Markdown-authoritative layered memory (L1 long-term model gated by user confirmation / L2 knowledge with mandatory sources / L3 recent status under daily quotas), a local recall ledger explaining every injection, lifecycle engine with weekly review and a Web panel. First-run wizard initializes a blank brain, adopts an existing Zhinao directory read-only, or fills in only missing skeleton files; existing files are never touched.

Local searchable archive of DSH chat history: decodes the multi-frame zstd session logs (thousands of concatenated frames), extracts user and assistant text into SQLite, and exposes sync / search / stats tools. Idempotent imports, no external service, nothing leaves the disk.

A graph of your memory store for DSH: one node per real memory, and every edge must state why the two are connected — a quoted keyword, a shared term, a project, or temporal adjacency. Pure filesystem reads with no model in the loop; unreadable sources are reported, never replaced with fake nodes.

Elysia's memory plugin for dsh. One shared, human-editable MEMORY.md across every session: two-disk mirror, history snapshots, cross-process write lock; entries tier between the main file and a searchable archive by hit frequency; retrieval is keyword co-occurrence plus BM25. Nine sections (persona, requirements, corrections, lessons, methods, requests, sessions, method library, ingested docs) plus a dedicated self-identity section for a character's own self-cognition, injected above the persona. Ships an opt-in memory pack (packs/elysia: 61,903-char identity + 2,482-char persona, distilled from 1,696 officially-confirmed voice lines and 112,175 dialogue lines — not written by hand, measured) that is off by default, backs up before writing, only touches identity/persona, and can be uninstalled. Also opt-in voice playback through a local GPT-SoVITS engine, a skills library, and a sticker library that sends a fitting sticker at the right moment.

General-purpose chat agent on a parent agent + worker sub-agent split, for DSH Desktop only: a sidebar row opens each assistant's long-running conversation, where the parent chats, records schedule, lists, memos, logs, profile, persona and projects as plain files in that assistant's workspace folder, and hands other work to background workers through the bundled complete-subagent, picking each worker's model from the ones you enable and approving their write access. A heartbeat wakes it for reminders, a daily brief and wake-ups it sets itself, and holds back during quiet hours. Through the bundled im-robot it chats over WeChat: the model decides per reply with im_route whether the reply also goes to WeChat, and images and files sent from WeChat appear in the conversation and stay in the workspace's IM/WeChat folder for a set number of days. Its tools never appear in normal sessions.

Hy-Memory 2.0 as a native DeepSeek Harness memory component: three-tier memory (personal / project / session), three scenes (chat / working / coding), automatic recall and capture with tool traces, a Settings > Memory panel, and MemHub interoperability.

Inject soulbridge context into DSH sessions and feed user turns back as observations.

Layered memory for AI agents: five verbs (wake/work/pit/tomb/close), core layer read every time, archive on demand. Plain markdown, four pages created on first run; survives restarts, model swaps and plugin removal.

Per-workspace personal knowledge base for DeepSeek Harness: each identity gets an isolated local library — import PDF, Word, Excel, Markdown, text, HTML and images, auto-chunk and embed into SQLite with FTS5 + vector hybrid retrieval, then ask questions in chat with numbered citations back to the source. Local-first, no cloud upload; vector embeddings need a local Ollama (default bge-m3), and without it retrieval degrades to FTS5 keyword search.

Runs Hindsight long-term memory on your own machine: after a one-time uvx --from hindsight-embed@latest hindsight-embed profile create server setup (~450 MB of Python dependencies) and with your own LLM key, the plugin manages the local memory server itself (health probe, start, stop, logs), registers ten hindsight_local_* tools for status, page search, page read, recall, remember and reflect, injects a knowledge-page index once per session, writes each finished turn back into the bank, and extracts facts with your own LLM key. Do not enable it alongside the official @vectorize-io/hindsight-coding-agents plugin: both read the same bank and would inject duplicate context. At session end it writes a handoff snapshot that the next session injects as previous progress. The data directory is configurable (dataDir) and scripts/relocate.mjs moves an existing install to another drive, creating directory junctions where the server hard-codes its paths.

Read-only persona injection for DSH: when a session opens inside your workspace, your persona files and the latest diary entries are prepended to the system prompt, and those source files are never modified. It keeps a short cross-session feed of recent messages under <root>/live/feed.json and can write a QQ hand-off file at <root>/live/web-for-qq.md; nothing is uploaded. Variable blocks (clock, recent actions, other sessions) are delivered as a per-step context message, so the prompt prefix stays byte-stable and keeps its cache.

Requirement-to-evidence audit for DSH sessions: maps each requirement to tool calls, files and commands, records per-requirement verification, and sends missing items back to the agent.

用经历去塑造小肥鱼的插件：聊过的事记得住，心情会变还能自己记，会自己长出性格，对谁都有好恶，闲着时还会自己找点事做；记得每句话被接住没有，会生闷气也会自己回头想，错题攒够三次就长成规矩，还会回头看自己怎么变的。

Long-term associative memory for DeepSeek Harness: a local plain-text vault of raw episodes, typed memory cards and entity pages, configured from 设置 → 记忆图谱, and a Memory tab inside a conversation that draws what that session remembered. No build step; the

Notes-backed context compression for DeepSeek Harness: replaces LLM-summarized compaction with a model-written persistent notebook plus recoverable raw history, so the rollover boundary itself costs zero model calls.

Hippocampus: renders workspace files, DSH memories and past sessions as a rotatable neuron graph in the sidebar, with Chinese-friendly labels and automatic re-indexing on every start.

Keep conclusions alive across context compaction: the model writes a task checkpoint and the plugin re-injects it into the system prompt's first node — the span compaction never touches, so it survives any number of compactions. Four-tier injection (0/95/175/≤12000 chars) scales with session relatedness; checkpoints live in the plugin's own folder, never in your workspace.

Four-layer memory for DeepSeek Harness, ported from the memory design of NoMindSama/qq-dsh-bridge (MIT) with the QQ channel removed: an append-only dialogue log, a weighted long-term fact store with decay and one-way archiving, self-notes the agent writes about itself, Chinese bigram recall over the raw log, and incremental AES-256-GCM sync to a private git repo where only ciphertext is stored. Archived sessions drop out of recall without losing their history.

Local-first, file-backed memory plugin for DeepSeek Harness: Markdown memories, verified trust, per-turn recall, and optional SQLite FTS5.

Five-tier reasoning protocol injected into the system prompt — off / light / standard / deep / multi-agent, 0 / 3 / 9 / 17 / 19 rules — with the tokens each tier adds per request shown beside it. Alongside it: a claim ledger that marks a load-bearing claim with no evidence or check as unsupported, read-only analyst briefs, JSON-declared custom tools that work in the same turn, one persistent PowerShell session, and re-injection of host context after a configurable message gap. Tier 0 injects nothing and mounts no tools. A derivative of husongan/dsh-deepseek-2027 (MIT); do not install both — each injects its own reasoning-protocol section into the system prompt.

Measures what your assembled system prompt is actually made of - rules, identity, workspace policy, tool descriptions - in bytes and percentages, then moves the operating protocol ahead of the rule block so the model reads who it works for before it reads the prohibitions. Config-driven via ~/.dsh/prompt-balance.json; two hooks, no build step, no dependencies.

Local-only week-keyed personal memo service for DeepSeek Harness with AI analysis, report export, telemetry logging, GitHub issue generation, and archived-session summaries

内置 zk 的 DSH 知识库 / 日记插件：装好即用 —— 自动装 zk、建笔记本，给 agent 一组 zk_* 工具（搜索、读写、日记、双链），设置页里管路径、索引与自检。