Skip to content
Agentmart
Submit a tool

Memtether

Cross-client AI memory hub: agents share one SQLite memory.db via file-level pointers.

Connector by lanbass869-cell · GitHub account verified by the MCP Registry

IndexedApache-2.0Local (stdio)★ 2
Write a review

Install

Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/lanbass869-cell-memtether

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Cross-client AI memory hub: agents share one SQLite memory.db via file-level pointers.

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list and install config. Checked Oct 25, 2026.

Tools
4 read · 2 write · 0 delete · 4 not yet classified
Secrets
None asked for
Install scripts
Not scanned yet
Runs as
Python package

Tools by what they do 10

Read from the source code; the running server may list different tools.

Read 4 · Look things up without changing anything.

  • as_of_query
  • list_memories
  • retire_memory
  • search_memory

Write 2 · Create, change, send or run things.

  • add_memories
  • pin_memory

Not yet classified 4 · We could not tell from the name and description.

  • correct_memory
  • feedback
  • profile
  • timeline

Secrets and settings it asks for

Other settings (environment variables or headers).

  • MEM_DB
  • MEM_DEFAULT_SOURCE

Files it touches

  • Any path the agent passes to its tools

Tools (10)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

  • add_memories

    写入一条记忆到共享记忆中枢（真源 memory.db，两个 WorkBuddy 版本与所有 MCP 客户端共用）。注：经 MCP 写入的记忆固定为 scope=shared；private/restricted 仅 CLI/API 可用。改记忆只走这里，禁止直接改 MEMORY.md / sink.json。重要：第一句必须把结论说完，后续 rebuild 生成导航投影时只取首句。

  • as_of_query

    双时间轴 as-of 查询：『在某个时间点，什么是真的（T轴）/系统知道什么（T'轴）』。kind=valid 用 valid_from/valid_to；kind=known 用 recorded_at/invalidated_at。

  • correct_memory

    纠正一条记忆：旧事实被 superseded（保留历史，不物理删除），新事实转为 active。用于发现记忆内容错误时的显式更正；会自动给旧事实记负向 Q-Value。

  • feedback

    回写 Q-Value 信号：告诉中枢某条记忆是否被采纳。

  • list_memories

    列出最近的 active 记忆（默认 20 条，仅首句摘要）。用于快速浏览中枢里有什么。

  • pin_memory

    钉住/取消钉住一条记忆（tags 加/去 pin）。被钉住的条目无条件进入投影，不参与『最近 N 条』时间竞争。不带 uid 调用 = 列出当前全部已钉住条目。

  • profile

    把某个 subject（用户/agent/项目）的 active 记忆聚合成一份**确定性表征**，可直接插入 system prompt。零 LLM：同一份库永远得到同一份结果，可逐条回溯 uid。用于回答『关于这个用户我已经知道什么』。

  • retire_memory

    退役一条记忆/工具：status→retired，双时间轴记账（valid_to+invalidated_at=现在）。用于确认某条记忆/资产已过时失效；会自动给该条记负向 Q-Value。

  • search_memory

    检索记忆中枢。返回结果是摘要片段，取全文请记下 uid 后再用本工具按关键词细查，或直接在 memory_hub 目录执行：memtether search "<关键词>"。下全称否定结论前必须先搜一次。

  • timeline

    沿替代链（superseded_by）还原一条记忆的完整演化史（最多 20 跳）。用于回答『这条结论是怎么变的』。

Signals

Stars
2
Forks
2
Open issues
4
Weekly downloads
574
Last commit
Oct 22, 2026 (7 days ago)
Last release
Oct 18, 2026 (12 days ago)
Last package publish
Oct 20, 2026 (9 days ago)

From GitHub (fetched Oct 24, 2026) and the package registry.

Details

Registry name
io.github.lanbass869-cell/memtether
Version
0.1.0a81
Auth
No auth
Packages
pypi memtether
License
Apache-2.0
Repository
github.com/MemTether/MemTether
First seen
Oct 24, 2026
Last checked
Oct 24, 2026
Source
Official MCP Registrygithub-graphql
Cite this page

Memtether MCP server. Agent Mart. https://agentmart.dev/connectors/lanbass869-cell-memtether (last updated 2026-10-24)