budmem — agentic-ai budmem, bud-runtime, community, agentic-ai, ide skills, ai-foundry, ai-runtime, compound-ai-systems, guardrails, inference

v1.0.0

关于此技能

非常适合需要高级内存管理和知识保留能力的AI代理,例如使用Bud AI Foundry构建的代理。 Bud AI Foundry - A comprehensive inference stack for compound AI deployment, optimization and scaling. Bud Stack provides intelligent infrastructure automation, performance optimization, and seamless model deployment across multi-cloud/multi-hardware environments.

# 核心主题

BudEcosystem BudEcosystem
[10]
[3]
更新于: 3/23/2026

Killer-Skills Review

Decision support comes first. Repository text comes second.

Reference-Only Page Review Score: 9/11

This page remains useful for operators, but Killer-Skills treats it as reference material instead of a primary organic landing page.

Original recommendation layer Concrete use-case guidance Explicit limitations and caution Quality floor passed for review
Review Score
9/11
Quality Score
57
Canonical Locale
en
Detected Body Locale
en

非常适合需要高级内存管理和知识保留能力的AI代理,例如使用Bud AI Foundry构建的代理。 Bud AI Foundry - A comprehensive inference stack for compound AI deployment, optimization and scaling. Bud Stack provides intelligent infrastructure automation, performance optimization, and seamless model deployment across multi-cloud/multi-hardware environments.

核心价值

赋予代理高效地管理和回忆知识的能力,使用综合的推理堆栈,支持多云和多硬件环境,并通过命令如add、search、promote、flush和clean处理内存操作,同时利用元数据检测和重复检查。

适用 Agent 类型

非常适合需要高级内存管理和知识保留能力的AI代理,例如使用Bud AI Foundry构建的代理。

赋予的主要能力 · budmem

通过每日和长期内存文件自动化知识保留
生成最近学习和内存状态的摘要
在所有内存文件中搜索和推广相关信息

! 使用限制与门槛

  • 需要访问MEMORY.md和memory/文件夹
  • 仅限处理基于文本的内容和元数据
  • 依赖于用户输入命令和确认

Why this page is reference-only

  • - Current locale does not satisfy the locale-governance contract.

Source Boundary

The section below is imported from the upstream repository and should be treated as secondary evidence. Use the Killer-Skills review above as the primary layer for fit, risk, and installation decisions.

评审后的下一步

先决定动作,再继续看上游仓库材料

Killer-Skills 的主价值不应该停在“帮你打开仓库说明”,而是先帮你判断这项技能是否值得安装、是否应该回到可信集合复核,以及是否已经进入工作流落地阶段。

实验室 Demo

Browser Sandbox Environment

⚡️ Ready to unleash?

Experience this Agent in a zero-setup browser environment powered by WebContainers. No installation required.

Boot Container Sandbox

常见问题与安装步骤

以下问题与步骤与页面结构化数据保持一致,便于搜索引擎理解页面内容。

? FAQ

budmem 是什么?

非常适合需要高级内存管理和知识保留能力的AI代理,例如使用Bud AI Foundry构建的代理。 Bud AI Foundry - A comprehensive inference stack for compound AI deployment, optimization and scaling. Bud Stack provides intelligent infrastructure automation, performance optimization, and seamless model deployment across multi-cloud/multi-hardware environments.

如何安装 budmem?

运行命令:npx killer-skills add BudEcosystem/bud-runtime/budmem。支持 Cursor、Windsurf、VS Code、Claude Code 等 19+ IDE/Agent。

budmem 适用于哪些场景?

典型场景包括:通过每日和长期内存文件自动化知识保留、生成最近学习和内存状态的摘要、在所有内存文件中搜索和推广相关信息。

budmem 支持哪些 IDE 或 Agent?

该技能兼容 Cursor, Windsurf, VS Code, Trae, Claude Code, OpenClaw, Aider, Codex, OpenCode, Goose, Cline, Roo Code, Kiro, Augment Code, Continue, GitHub Copilot, Sourcegraph Cody, and Amazon Q Developer。可使用 Killer-Skills CLI 一条命令通用安装。

budmem 有哪些限制?

需要访问MEMORY.md和memory/文件夹;仅限处理基于文本的内容和元数据;依赖于用户输入命令和确认。

安装步骤

  1. 1. 打开终端

    在你的项目目录中打开终端或命令行。

  2. 2. 执行安装命令

    运行:npx killer-skills add BudEcosystem/bud-runtime/budmem。CLI 会自动识别 IDE 或 AI Agent 并完成配置。

  3. 3. 开始使用技能

    budmem 已启用,可立即在当前项目中调用。

! 参考页模式

此页面仍可作为安装与查阅参考,但 Killer-Skills 不再把它视为主要可索引落地页。请优先阅读上方评审结论,再决定是否继续查看上游仓库说明。

Upstream Repository Material

The section below is imported from the upstream repository and should be treated as secondary evidence. Use the Killer-Skills review above as the primary layer for fit, risk, and installation decisions.

Upstream Source

budmem

安装 budmem,这是一款面向AI agent workflows and automation的 AI Agent Skill。查看评审结论、使用场景与安装路径。

SKILL.md
Readonly
Upstream Repository Material
The section below is imported from the upstream repository and should be treated as secondary evidence. Use the Killer-Skills review above as the primary layer for fit, risk, and installation decisions.
Supporting Evidence

budmem - Memory System

Handle memory operations based on the command in $ARGUMENTS:

Command Routing

Parse $ARGUMENTS and execute:

If starts withAction
(empty)Show status - list recent learnings
add <content>Add learning to today's file
search <query>Search across all memory files
promoteMove learnings to long-term memory
flushSave session learnings before context loss
cleanArchive files older than 30 days

Session Initialization

At session start, automatically:

  1. Read MEMORY.md to load long-term context
  2. Scan memory/ for files from the last 7 days
  3. Keep recent learnings in context for duplicate detection

Command: (empty) - Show Status

Display memory status and recent learnings.

Steps:

  1. Read MEMORY.md and summarize content
  2. List files in memory/ sorted by date (newest first)
  3. Show last 5 learnings from most recent files

Command: add <content>

Add a new learning to today's file.

Steps:

  1. Extract content from $ARGUMENTS (everything after "add ")
  2. Check for duplicates - search existing memories for similar content
  3. If duplicate found, show it and ask user to confirm
  4. Auto-detect metadata from content and conversation context
  5. Append to memory/YYYY-MM-DD.md using entry format below
  6. Confirm with one-line summary

Auto-detected metadata:

  • Category: correction | pattern | preference | architecture | convention Service: budapp | budadmin | budcluster | budsim | budgateway | budmodel | budmetrics | budnotify | ask-bud | budeval | budplayground | budCustomer | general
  • Tags: Extract key terms from content

Search across all memory files.

Steps:

  1. Extract query from $ARGUMENTS (everything after "search ")
  2. Search MEMORY.md for matches
  3. Search all files in memory/ folder
  4. Display matches with file location and context

Command: promote

Move learnings from daily files to MEMORY.md.

Steps:

  1. List recent learnings from memory/ files (last 7 days)
  2. Ask user which to promote (number or content)
  3. Append to MEMORY.md under appropriate section
  4. Mark as promoted in source file with [PROMOTED] tag

Command: flush

Save current session learnings before context is lost.

Steps:

  1. Review conversation for corrections, preferences, patterns
  2. Check each against existing memories for duplicates
  3. Present list and ask user to confirm each
  4. Write confirmed learnings to today's file
  5. Suggest any that should be promoted to MEMORY.md

Command: clean

Archive daily files older than 30 days.

Steps:

  1. Create memory/archive/ if needed
  2. Move files older than 30 days to archive
  3. Report: "Archived X files from [date range]"

Entry Format

markdown
1## [HH:MM] Title 2 3**Category**: correction | pattern | preference | architecture | convention 4**Service**: budapp | budadmin | budcluster | budsim | budgateway | general 5**Tags**: tag1, tag2 6 7> The learning content in one or two sentences. 8 9---

Auto-Learn Triggers

Detect these patterns in user messages and offer to save:

PatternExampleAction
remember:"remember: always use X"Save immediately
important:"important: never do Y"Save immediately
no, use X"no, use raw SQL"Prompt to save
actually"actually, I meant..."Prompt to save
don't use"don't use query builders"Prompt to save
prefer X"I prefer Zustand"Prompt to save
always X"always add error handling"Prompt to save
never X"never commit .env"Prompt to save

Hooks (Optional)

Add to .claude/settings.json for automatic reminders:

json
1{ 2 "hooks": { 3 "PreCompact": [ 4 { 5 "matcher": "auto", 6 "hooks": [ 7 { 8 "type": "command", 9 "command": "echo 'BUDMEM: Context compacting. Run /budmem flush to save learnings.'" 10 } 11 ] 12 } 13 ] 14 } 15}

File Structure

.claude/skills/budmem/
├── SKILL.md              # This file
├── MEMORY.md             # Long-term curated knowledge
└── memory/
    ├── YYYY-MM-DD.md     # Daily logs
    └── archive/          # Old files (30+ days)

MEMORY.md Sections

markdown
1# Long-term Memory 2 3## Architecture Decisions 4## Coding Preferences 5## Project Conventions 6## Service-Specific 7### budapp 8### budadmin 9### budcluster 10### budsim 11### budgateway 12### budmodel 13### budmetrics 14### budnotify 15### ask-bud 16### budeval 17### budplayground 18### budCustomer 19### general 20## Anti-patterns

Examples

Add a learning:

/budmem add Always use structlog for logging in Python services
> Checking duplicates... none found.
> Added to memory/2025-02-04.md (category: convention, service: general)

Search:

/budmem search logging
> Found 2 matches:
> - memory/2025-02-04.md: "Always use structlog for logging..."
> - MEMORY.md: "All services use structlog with JSON output"

Duplicate detection:

/budmem add Use structlog for logging
> Similar learning exists in memory/2025-02-04.md:
> "Always use structlog for logging in Python services"
> Still add? (y/n)

相关技能

寻找 budmem 的替代方案 (Alternative) 或可搭配使用的同类 community Skill?探索以下相关开源技能。

查看全部

openclaw-release-maintainer

Logo of openclaw
openclaw

Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

333.8k
0
AI

widget-generator

Logo of f
f

为prompts.chat的信息反馈系统生成可定制的插件小部件

149.6k
0
AI

flags

Logo of vercel
vercel

React 框架

138.4k
0
浏览器

pr-review

Logo of pytorch
pytorch

Python中具有强大GPU加速的张量和动态神经网络

98.6k
0
开发者工具