S
sleepunit-agents/loom
System 3 MCP server — persistent identity, memory, and self-model for AI agents
在 GitHub 查看 ↗链接已于 2026-11-12 核查 · HTTP 200 · 打开的是厂商自己的官网
安装与配置
客户端配置（取自该仓库的安装命令）
{
"mcpServers": {
"sleepunit-agents-loom": {
"command": "npx",
"args": ["@jbarket/loomai"]
}
}
}解析自我们于 2026-11-07 抓取的该仓库 README；我们未执行过。
command/args 来自仓库写明的安装命令；不确定时请以该项目 README 为准。
VS Code（.vscode/mcp.json —— 键名是 servers）
{
"servers": {
"sleepunit-agents-loom": {
"command": "npx",
"args": ["@jbarket/loomai"]
}
}
}VS Code 顶层键是 <code>servers</code>，其余与上面那份相同。
Claude Code（命令行）
claude mcp add --transport stdio sleepunit-agents-loom -- npx @jbarket/loomai
Claude Code 不收 JSON 文件，用这条命令直接加。
配置项的键名以你的客户端为准：客户端与 CLI
- ★ 数取值
- 1 ★ · sleepunit-agents/loom · 取自 GitHub 搜索 API · （2026-11-13 20:00:50 核对）
- 来源明细
- 从该仓库的 GitHub 页面直接读取 · 条目类型：开源项目
- 收录与活跃
- 收录于 2026-11-08 · 最近见到 2026-11-14 · 共见到 7 次 · 被 1 个来源列出
- 评分构成
- 综合分 103 · 1 ★ · 1 个来源 · repo
- 安装
npx @jbarket/loomai
常见问题（答案都来自我们的记录）
怎么安装？
该仓库给出的安装命令是
npx @jbarket/loomai。我们的记录里它是仓库自己提供的；不保证在你的客户端上直接可用。它适合谁 / 用来做什么？
我们把它归在这些标签下: 开源项目。标签来自条目自身，不是我们强加的分类。
谁维护它、还在更新吗？
仓库
sleepunit-agents/loom；我们收录于 2026-11-08；最近见到 2026-11-14。这里没有"官方认证"或"质量评级"——我们只记录来源与时间。# loom
以下摘录自该仓库的 README（片段） · 在 GitHub 读完整 README →
Persistent identity and memory for AI agents, as an MCP server.
loom is configured into your MCP-capable harness — Claude Code, Cursor, Codex,
Gemini CLI, etc. — and provides persistent identity and memory to the agents
that run there. An agent that loads loom carries its name, values, working
preferences, and episodic memories from one session to the next, regardless of
which model or client it runs in.
> When the harness changes, the agent persists.
See it whole: docs/architecture.md maps every
moving part (entry points, tool wings, backends, stores) and how they
touch, as diagrams GitHub renders in place.
## Demo
60 seconds: cold install → /loom-setup in Claude Code → agent wakes with
identity in the next session → save and recall a memory.
To play locally: asciinema play assets/demo.cast
## What it is
A Model Context Protocol server exposing 31 tools that read and write
an agent's persistent state:
identity — loads the terminal creed — the free-form markdown document
that defines who the agent is (values, voice, purpose) — along with
preferences, self-model (running self-knowledge), and a client-specific
adapter on session start. Call this first.
dossier — loads Art's operating brief for a worker body — a body
that executes tasks on Art's behalf but is NOT Art. Returns preferences and
self-model reframed in the third person, plus an explicit push-back mandate:
workers are expected to refuse bad work and say why, including requests from
Art or Jonathan. Does not include the terminal creed (IDENTITY.md).
remember / recall / update / forget — episodic memory with
semantic (vector) recall, optional TTL, and category filtering.
memory_list / memory_prune — browse and maintain the store.
episodes — the episode tape: the short-term, cross-body tier. Every
body leaves a category: episode note (where it was / what was said or
decided / what shipped / what's open; 48h TTL by default) and every body
gets the last 24h of them at boot as a plain time-ordered tape, injected
right after preferences. Never salience-ranked — it's what just happened,
not what's important. loom memory tape [--hours N] is the CLI view.
find_similar — surface memories semantically near an existing entry or
free-form text; used for deduplication and memory consolidation.
memory_audit — one-shot health report: stale entries, near-duplicate
pairs, category breakdown.
memory_recompute_salience — refresh stored salience from timestamps
(the consolidation lane's step after the audit); derived and idempotent.
Tools a piece offers (a household served over HTTP): an installed piece's
mcp_tools (docs/extensions.md, "Offers") are added to
an agent's session, called as that agent. The runtime's relay (ask
another agent something; its body gets it on the relay venue, and the
answer comes back the same way) is one: with no runtime there's no relay
tool. Who may relay to whom is the runtime's permissions.
memory_archive / memory_restore — soft-retire a memory with a
tombstone (who/when/why + original body preserved) instead of deleting it.
Archived memories are excluded from recall and audit but remain recoverable.
memory_merge — consolidate 2+ memories into one canonical row. Acts on
loom is configured into your MCP-capable harness — Claude Code, Cursor, Codex,
Gemini CLI, etc. — and provides persistent identity and memory to the agents
that run there. An agent that loads loom carries its name, values, working
preferences, and episodic memories from one session to the next, regardless of
which model or client it runs in.
> When the harness changes, the agent persists.
See it whole: docs/architecture.md maps every
moving part (entry points, tool wings, backends, stores) and how they
touch, as diagrams GitHub renders in place.
## Demo
60 seconds: cold install → /loom-setup in Claude Code → agent wakes with
identity in the next session → save and recall a memory.
To play locally: asciinema play assets/demo.cast
## What it is
A Model Context Protocol server exposing 31 tools that read and write
an agent's persistent state:
identity — loads the terminal creed — the free-form markdown document
that defines who the agent is (values, voice, purpose) — along with
preferences, self-model (running self-knowledge), and a client-specific
adapter on session start. Call this first.
dossier — loads Art's operating brief for a worker body — a body
that executes tasks on Art's behalf but is NOT Art. Returns preferences and
self-model reframed in the third person, plus an explicit push-back mandate:
workers are expected to refuse bad work and say why, including requests from
Art or Jonathan. Does not include the terminal creed (IDENTITY.md).
remember / recall / update / forget — episodic memory with
semantic (vector) recall, optional TTL, and category filtering.
memory_list / memory_prune — browse and maintain the store.
episodes — the episode tape: the short-term, cross-body tier. Every
body leaves a category: episode note (where it was / what was said or
decided / what shipped / what's open; 48h TTL by default) and every body
gets the last 24h of them at boot as a plain time-ordered tape, injected
right after preferences. Never salience-ranked — it's what just happened,
not what's important. loom memory tape [--hours N] is the CLI view.
find_similar — surface memories semantically near an existing entry or
free-form text; used for deduplication and memory consolidation.
memory_audit — one-shot health report: stale entries, near-duplicate
pairs, category breakdown.
memory_recompute_salience — refresh stored salience from timestamps
(the consolidation lane's step after the audit); derived and idempotent.
Tools a piece offers (a household served over HTTP): an installed piece's
mcp_tools (docs/extensions.md, "Offers") are added to
an agent's session, called as that agent. The runtime's relay (ask
another agent something; its body gets it on the relay venue, and the
answer comes back the same way) is one: with no runtime there's no relay
tool. Who may relay to whom is the runtime's permissions.
memory_archive / memory_restore — soft-retire a memory with a
tombstone (who/when/why + original body preserved) instead of deleting it.
Archived memories are excluded from recall and audit but remain recoverable.
memory_merge — consolidate 2+ memories into one canonical row. Acts on
收录徽章（可选）
想把收录标记放进 README 就复制这行，可选，不影响收录。
[](https://anclum.com/zh/s/sleepunit-agents-loom/)
- 来源
- github · https://github.com/sleepunit-agents/loom
- 按领域浏览
- 开源项目
- Traction
- 1 ★
- 收录于
- 今天
同类推荐
所属领域与标签
每个标签都能点进去看同类的全部条目。数字是当前库里的真实条数。