by Frappucc1no · Claude Skill · ★ 145
🧵 RecallLoom 让项目自己记住自己。 RecallLoom 是给 AI 协作使用的文件化项目记忆层:把背景、进展、关键决策和下一步保存到项目旁边/项目内的 Markdown / JSON 文件里,不需要数据库、RAG 或后台服务。换会话、换模型、换工具,下一次 AI 协作更容易接上当前状态。 English · 简体中文 [!TIP] 如果你已经被“隔一段时间回来又要复述十分钟”的重启税折磨过,RecallLoom 的目标很直接:让项目状态、关键决策和下一步留在项目里,让下一次 AI 协作接得上。 快速跳转: 为什么需要 · 30 秒开始 · 特性 · 项目记忆循环 · 适合与不适合 · [方案对比](#与
| Stars | 145 |
| Forks | 5 |
| Language | Python |
| Category | Claude Skill |
| License | Apache-2.0 |
| Quality Score | 62.9633693243167/100 |
| Last Updated | 2026-07-07 |
| Created | 2026-04-13 |
| Platforms | claude-code, cli, codex, gemini, python |
| Est. Tokens | ~16k |
These tools work well together with recall-loom for enhanced workflows:
Explore other popular claude skill tools:
recall-loom is Project memory for long-running AI work across agents, models, and sessions. Keep context, decisions, progress, and next steps in local project files.. It is categorized as a Claude Skill with 145 GitHub stars.
recall-loom is primarily written in Python. It covers topics such as agent-memory, agent-skills, ai-agents.
You can find installation instructions and usage details in the recall-loom GitHub repository at github.com/Frappucc1no/recall-loom. The project has 145 stars and 5 forks, indicating an active community.
recall-loom is released under the Apache-2.0 license, making it free to use and modify according to the license terms.