独立运营 · 不出售排名位置  · 我们怎么评估工具 →
Anclum 提交收录
S

Sigillion-AI/sigillion

Lossless, local, provable memory for AI agents: every Claude Code session kept on your own computer, word for…

装之前:没有可直接复制的命令 · 许可证 NOASSERTION · 链接可访问

第一步:Windows: py -m pip install --user pipx, then py -m pipx ensurepath.

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
★ 数取值
29 ★ · Sigillion-AI/sigillion · 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 2026-11-11 · 最近见到 2026-11-14 · 共见到 4 次 · 被 1 个来源列出
评分构成
综合分 103 · 29 ★ · 1 个来源 · repo
常见问题(答案都来自我们的记录)
它适合谁 / 用来做什么?
我们把它归在这些标签下: 记忆增强。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 Sigillion-AI/sigillion;我们收录于 2026-11-11;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。

# Sigillion

以下摘录自该仓库的 README(片段) · 在 GitHub 读完整 README →

> Remember everything. Read almost nothing. Prove all of it.
Website | PyPI | Releases | Changelog | Privacy | Terms | Security | For agents
Sigillion is memory for AI agents, made by Steddy Nova SRL. It keeps the whole record of what an agent and a person did,
word for word, on your own computer; compressed, so that a model can skim weeks of work in a few hundred tokens and open
the full text only where it needs it; sealed, so that every record can be checked to show it was not altered. No
telemetry: Sigillion itself sends nothing anywhere. What it hands the agent becomes part of your session, which Claude
Code sends to Anthropic and Cursor to the models it uses, as each does with the rest.
## One engine, in order
1. Claude Code and Cursor first. Today Sigillion works in Claude Code and in Cursor: every session is kept after
compaction and after Claude Code deletes its transcripts. Claude searches and reopens its own history while it works.
Both agents keep their sessions in one store on your computer: in the same project, Cursor's agent can recall what
Claude Code did. Claude can recall what you did in Cursor. Claude Code is where every feature is built, measured
and proved on real work. It is the first step, not the whole idea.
2. Every AI model and agent tool next. The engine does not depend on one model. The same sealed memory will sit under
other models and the tools people already use (coding agents, agent frameworks, assistants), with an SDK for anyone
who builds agents.
3. Then the fields where the record must stand: legal and finance. The same store, with a pack for each field: its
citations and terms, its redaction rules and receipts that show what the model did and when, for work a client, a
regulator or a court may ask about.
Each step comes after the one before has proved itself with other people's numbers, not before.
## Install
Sigillion 0.1.3 is on PyPI. It fixes sessions that could stop saving at a Windows path. 0.1.2 added the Mac with
Apple silicon; 0.1.1 added Cursor and took the trial out. What each release changed: CHANGELOG.md.
You need:
Windows or Linux on a 64-bit Intel or AMD processor (x86_64); or a Mac with Apple silicon (M1 or later) and macOS 11
or later. Not yet: a Mac with an Intel processor, Windows on ARM, ARM Linux, Alpine Linux.
Python 3.10 to 3.14. Python 3.15 needs a C compiler until zstandard, which Sigillion uses, publishes a package for
it; with 3.14 installed beside it, pipx install --include-deps --python 3.14 sigillion needs none (on Linux with
a pipx older than 1.6: --python python3.14).
Claude Code 2.1.287 or later. To update it: claude update; installed with WinGet:
winget upgrade Anthropic.ClaudeCode; installed with Homebrew: brew upgrade claude-code.
Or Cursor (measured on Cursor 3.27 on Windows).
pipx, then a new terminal:
Windows: py -m pip install --user pipx, then py -m pipx ensurepath.
Ubuntu 23.04 or later, Debian 12 or later: sudo apt install pipx, then pipx ensurepath.
Other Linux: python3 -m pip install --user pipx, then python3 -m pipx ensurepath.
macOS: brew install pipx, then pipx ensurepath.
Anything else: pipx's installation page.
Then the commands:
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。
[![Listed on Anclum](https://anclum.com/badge/listed.svg)](https://anclum.com/zh/s/sigillion-ai-sigillion/)
来源
github · https://github.com/Sigillion-AI/sigillion
按领域浏览
记忆增强
Traction
29 ★
收录于
今天

同类推荐

全部 48 →

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。

浏览全部领域 →

← 返回索引