engram
Skill1 starsChanged 30 days agoMITView on GitHub
Self-evolving global memory for Claude Code, shared across projects. Single entry with subcommands - setup to enable or disable automatic memory by registering/removing SessionStart and Stop hooks in user settings, review to audit and tidy memory entries, sync to sync memory across machines via a git remote, status to show hook registration and data state. Use when the user asks to set up, enable, check, tidy, review, back up, or sync memory, or mentions engram.
9 other projects use a version of this.
Adopted by
Most recently taken up by
- mazze93/claude-code-skillstheir own version, first seen Nov 1, 2026
- techtheist/engramtheir own version, first seen Oct 31, 2026
- JJHbrams/Project-AMBERtheir own version, first seen Oct 20, 2026
- LucasXCoder/engramtheir own version, first seen Oct 9, 2026
- poorants/engramtheir own version, first seen Sep 30, 2026
- ayvazyan10/engramtheir own version, first seen Aug 27, 2026
Other versions9 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 9
- Patdolitse/piia-engram★ 165same name
- EngramMemory/engram-memory★ 36same name
- techtheist/engram★ 34same name
- DazhuangJammy/Engram★ 18same name
- ayvazyan10/engram★ 10same name
- JJHbrams/Project-AMBER★ 2same name
- LucasXCoder/engram★ 1same name
- mazze93/claude-code-skills★ 1same name
- poorants/engram★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository31 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- agentdocs-setupSkill
- agent-docsSkill
- codebase-designSkill
- code-reviewSkill
- commit-msgSkill
- diagnoseSkill
- handoffSkill
- improve-archSkill
- interviewSkill
- prototypeSkill
- prSkill
- questionnaireSkill
- researchSkill
- review-frontend-productionSkill
- run-ticketSkill
- setup-gatesSkill
- sync-claude-mdSkill
- sync-readmeSkill
- task-retroSkill
- task-specSkill
- task-ticketsSkill
- tddSkill
- teachSkill
- team-devSkill
- wait-whatSkill
- wayfinderSkill
- which-skillSkill
- wizardSkill
- writing-for-agentsSkill
The file
Files
engram · 全局自进化记忆
数据目录 ${ENGRAM_MEMORY_DIR:-$HOME/.engram}(下称 $GM)中的有界记忆:
MEMORY.md(跨项目事实,≤3000 字符)、USER.md(用户画像,≤1500 字符)、archive/(淘汰条目)。
自动能力由本 skill 目录下 hooks/ 的两个脚本提供,经 /engram setup 登记进用户级 settings 后生效:
- 会话启动注入(SessionStart ·
hooks/inject.js):维护规则与记忆内容进入会话上下文; - 会话结束巩固(Stop ·
hooks/consolidate.js):后台子进程提炼本次会话,更新有界记忆。
分发
按 $ARGUMENTS 依序分支;命中子命令后读取本 skill 目录下对应文件执行,分发器不承载子流程细节:
- 空或
status→ 执行下方「状态与提示」; setup(enable / disable / repair 视为同义)→ 读取同目录setup.md执行;review或memory-review→ 读取同目录memory-review.md执行;sync→ 读取同目录sync.md执行;- 其他文本 → 按语义就近匹配,复述理解并经用户确认后进入对应流程;无法匹配时列出四个子命令及一句话说明。
状态与提示(status / 无参数)
- hooks 登记:读取
~/.claude/settings.json,检查hooks.SessionStart/hooks.Stop中是否存在指向本 skillhooks/的条目;未启用时提示运行/engram setup; - 数据目录:
$GM是否存在、MEMORY.md 与 USER.md 是否初始化及当前字符量(上限 3000 / 1500)、是否为 git 仓库、工作区是否干净; - 最后输出子命令提示:
/engram setup 启用或停用自动记忆(hooks 登记)
/engram review 审查、整理记忆条目
/engram sync 经 git 远程多机同步记忆
/engram status 查看登记与数据状态跨平台注意
- 数据目录:由 hooks 脚本经
os.homedir()与path.join计算——macOS/Linux 为~/.engram,Windows 为%USERPROFILE%\.engram;ENGRAM_MEMORY_DIR覆盖在各平台行为一致; - 文档记法:本文档及子流程文档中的
${ENGRAM_MEMORY_DIR:-$HOME/.engram}、git -C "$GM"、~/.claude/settings.json为 bash 系记法,macOS/Linux 原生适用;Windows 下经 Claude Code 的 Bash 工具(Git Bash)执行同样可用,若需在 PowerShell / cmd 中手动执行等效命令,改用对应 shell 的变量语法与路径写法; - 已登记的 hooks 无上述限制:settings.json 中的 command 为双引号包裹的绝对路径,bash / PowerShell / cmd 均按字面解析(详见
setup.md写入纪律)。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

