Y

yjh051108/@dsh-external/dsh-engram-relay

“大 engram 小 KV：外置 engram 转接层（跨会话分层记忆：global/project/session，AI 自主决策归属 + 因果链接索引 + 渐进披露 + 主动唤醒），为 DSH 实现超长上下文记忆稀疏路由，与官方 co…”

OverviewREADME

TL;DR

大 engram 小 KV：外置 engram 转接层（跨会话分层记忆：global/project/session，AI 自主决策归属 + 因果链接索引 + 渐进披露 + 主动唤醒），为 DSH 实现超长上下文记忆稀疏路由，与官方 compact 共存（官方腾 KV，engram 保细节）

30 days of GitHub stars

+0 stars

+0% observed change

Oct 19+0 stars3 totalOct 20+0 stars3 totalOct 21+0 stars (estimated)3 totalOct 22+0 stars (estimated)3 totalOct 23+0 stars (estimated)3 totalOct 24+0 stars (estimated)3 totalOct 25+0 stars (estimated)3 totalOct 26+0 stars (estimated)3 totalOct 27+0 stars (estimated)3 totalOct 28+0 stars (estimated)3 totalOct 29+0 stars (estimated)3 totalOct 30+0 stars (estimated)3 totalOct 31+0 stars (estimated)3 totalNov 1+0 stars (estimated)3 totalNov 2+0 stars (estimated)3 totalNov 3+0 stars (estimated)3 totalNov 4+0 stars (estimated)3 totalNov 5+0 stars (estimated)3 totalNov 6+0 stars (estimated)3 totalNov 7+0 stars (estimated)3 totalNov 8+0 stars (estimated)3 totalNov 9+0 stars (estimated)3 totalNov 10+0 stars3 totalNov 11+0 stars3 total

Oct 14Oct 28Nov 12

Also in this category

4 items

V

Memory

vectorize-io/hindsight

Hindsight, agent memory that learns: long-term project memory with auto recall and retain, knowledge pages, deep reflection, and per-repo memory banks.

★ 60.3K

V

Memory

volcengine/OpenViking

OpenViking memory and context bundle for DeepSeek Harness: pre-step auto-recall and profile injection, session capture, `viking://` URI guarding, and recall/write memory tools backed by an OpenViking server.

★ 42K

M

Memory

MemTensor/MemOS

Local-first memory plugin for DeepSeek Harness: layered L1/L2/L3 memory with automatic capture, manual memos_* retrieval tools, and a built-in memory viewer.

★ 12.4K

A

Memory

agentscope-ai/ReMe

Connects DeepSeek Harness to ReMe's local-first, self-evolving personal knowledge base: automatically captures completed main-agent conversations as user-owned Markdown memory, searches conversations and source material through reme_search with BM25, optional embeddings, and wikilink expansion, and schedules daily memory consolidation.

★ 3.8K