Turn your team's AI coding sessions, GitHub code, and docs into one shared, searchable memory — a self-hosted git truth store you query right from your editor over MCP.
Codebase Context gives AI agents understanding of your codebase through semantic code search, team conventions, patterns, and memory, so they use fewer tokens, spend less time, and produce better, more familiar output.
Snack Desktop：面向企业增长团队的 AI 工作台桌面端，把知识、会议、IM 和业务数据接进同一个工作流。
Project memory for coding agents: team-shareable engineering experience assets with Milvus semantic retrieval.
Watercooler: your team's shared reasoning layer for agentic coding. Preserve the "why" around your code.
Give every AI agent persistent memory of your team's knowledge. No vector DB, no RAG — just Git + BM25 + 114 tokens per session.
Self-evolving team memory
Git-native shared AI memory for teams. Your team's knowledge, conventions, and decisions — automatically loaded into Claude Code, Cursor, and Copilot.
Open-source team memory layer for AI coding agents — markdown files in git, user→team→org hierarchy, cross-vendor MCP server. Apache-2.0.
AI memory for teams. The Arroway plugin for Claude Code, Cowork and Cursor: your team writes down what it decided, and every teammate's AI reads it before it acts. Mirror of the plugin package.
给团队 AI 编程助手用的共享项目记忆：先确认再共享，来源可追溯，换机器也能接着干
Shared memory and switchboard for teams whose members each run their own AI agents. Meetings → decisions and action items, team knowledge graph, MCP tool surface, A2A broker with human oversight.
One shared memory for your whole company — plain Markdown in git that your agent loads at session start and your team reads in a browser.
The memory layer for your team. Mneme auto-extracts decisions, questions, and action items from AI chats — and pings you when they go stale
Team memory engine for AI agents — write, route, graph, recall, inject. Self-hostable.
Git-native shared development memory for Claude Code teams
Mention the bot in Slack and the coding agent on your own machine does the work — Claude Code, Codex, OpenCode, Cursor, over ACP. The answer streams back into the thread, destructive steps ask first, and team memory is scoped to the channel that asked. Zero open ports, your keys, your hardware.
Token-efficient shared project memory skills for developers and AI coding agents.
Memória compartilhada de time para o Claude Code: entrevista aboutme, encerrar trabalho, passagem de bastão e git pull no início da sessão.
To associate your repository with the team-memory topic, visit your repo's landing page and select "manage topics."