curated collections / memory-context
DSH Memory & Context Plugins (DeepSeek Harness)
Automated recall, project memory and context compression — the hottest direction in DSH right now.
The clearest signal in the DeepSeek Harness catalog: the top installable entries by stars are a memory stack — pre-step auto-recall, profile injection, session capture and project-level memory. These DSH plugins attack the real cost centers of long-running agents: context bloat, cross-task forgetting, and messy recall.
The strongest DeepSeek Harness (DSH) memory stack: auto-recall project memory first, then context compression — ranked below by stars and recency.
472 plugins106.4K stars
Synced from our GitHub list · Catalog updated 2026-09-05
Why it matters now
Highest-signal category — the most-starred installable entries (OpenViking-style memory, Hindsight-style project memory) are all memory.
Attacks real costs — deterministic compaction, cross-session memory and token savings map 1:1 to agent cost and confusion.
High ceiling — a quality memory layer is exactly what is hardest to copy vs. another theme or sidebar.
In this collection
Editor’s picks lead, then the rest of the theme is auto-discovered by GitHub topics and ranked by stars. Catalog data, refreshed on every sync.
- 1openvikingby volcengine
Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
Editor’s pickCuratedAgents, Automation & Workflows35.5K2,711 - 2memosby memtensor
Self-evolving memory OS for LLM & AI Agents: ultra-persistent memory, hybrid-retrieval, and cross-task skill reuse, with 35.24% token savings and DeepSeek Harness support.
Editor’s pickCuratedSessions & Messages11.2K1,022 - 3everosby evermind-ai
One portable memory layer for every AI agent: local-first, Markdown-native, user-owned, and self-evolving across apps, tools, and workflows.
Editor’s pickMCP & Protocols12.7K913 - 4mindmemosby mindscale-noah
Gives dsh agents persistent cross-session memory: auto-recalls relevant user/task context before each turn and writes back lessons after each turn, evolving memory via schema learning and skill distillation
Editor’s pickCuratedSessions & Messages97795 - 5mnemonby mnemon-dev
LLM-supervised persistent memory for AI agents — graph-based recall, cross-session knowledge, single binary. Works with DeepSeek Harness, Claude Code, OpenClaw, and any agent runtime.
Editor’s pickCLI & Terminal55668 - 6dsh-mnemonby omdsh-dev
Composable three-tier memory control plane for DeepSeek Harness: persistent runtime context, searchable project documents, pluggable long-term memory, guarded strategies, WebUI, and headless tools.
Editor’s pickCuratedSessions & Messages32921 - 7engramoryby tinqiao-oss
A portable memory protocol for AI agents — load it as standing rules; a curation discipline + reference spec + optional cap hook.
Editor’s pickCuratedSessions & Messages18611 - 8weknoraby tencent
Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, and a self-maintaining Wiki.
CuratedKnowledge & Research21.3K3,085 - 9remeby agentscope-ai
ReMe: Memory Management Kit for Agents - Remember Me, Refine Me.
CuratedKnowledge & Research3,410293 - 10memsearchby zilliztech
A persistent, unified memory layer for all your AI agents (e.g. Claude Code, Codex, DSH), backed by Markdown and Milvus.
CuratedKnowledge & Research2,565237 - 11graph-memoryby adoresever
Knowledge-graph memory core native to DSH (Cordis adapter, cross-session triples).
CuratedKnowledge & Research59288 - 12deja-vuby vshulcz
Memory for coding agents built from the session history already on disk: Claude Code, Codex, Cursor, Copilot CLI, OpenClaw and 17 more. Search months of past sessions, including everything from before you installed it, and recall them in any agent. No LLM, no embeddings, one local Go binary.
CuratedKnowledge & Research77364 - 14ai-novel-writerby ethanyoq
Installs a dedicated AI novel-writing preset and workbench: revisioned local project assets, a compact side drawer, and native approval-gated single-file changes.
CuratedAgents, Automation & Workflows61282 - 15memtrace-publicby syncable-dev
Structural memory for AI coding agents. Bi-temporal graph, MCP-native, zero LLM calls. Cursor · Claude Code · Codex · DeepSeek Harness · Hermes · VS Code · Windsurf.
CuratedKnowledge & Research46941 - 16flowixby text2future
Registers the local flowix-cli MCP server so the agent can search, read, create, and edit Flowix memos and mind-map artifacts.
CuratedAgents, Automation & Workflows39349 - 17vibe-skillsby foryourhealth111-pixel
Intelligent Skill routing and workflow orchestration for AI agents — +21.12 pp reward, −29.6% tokens on SkillsBench with DeepSeekV4Flash-VE.
Agents, Automation & Workflows3,156262 - 18memmy-agentby memtensor
🍙 A personal AI agent & local memory hub for all AI agents, gives every AI one shared, fully controlled memory and persistent context — all AI remember the same you. Now supports Claude Code, Codex, OpenClaw and Hermes Agent etc.
MCP & Protocols1,256111
Frequently asked questions
What does a memory plugin do in DeepSeek Harness?
It gives an agent long-term recall — storing notes, profiles or past turns — so it remembers context across sessions instead of starting fresh each time.
How does context compression help DSH agents?
It deterministically summarizes or truncates the conversation, shrinking token usage and preventing context bloat on long-running tasks.
What's the difference between profile memory and session memory?
Session memory lasts for one run; profile or project memory persists across runs and is injected each time so the agent keeps what it learned.
Which memory plugin should I start with?
Start with a project-memory plugin with auto-recall — the highest-signal install — then add compaction once your agents run long sessions.
Do memory plugins increase API costs?
They add a little overhead per recall, but usually lower total cost by avoiding re-reading large contexts and re-doing forgotten work.
Explore related categories
Explore the full directory
Hundreds of plugins across every category — or submit your own and get listed.