{"slug": "show-hn-agentconnect-shared-agents-with-separate-permissions", "title": "Show HN: AgentConnect, shared agents with separate permissions", "summary": "AgentConnect, an open-source platform for teams and multiple AI agents to collaborate across Slack, Telegram, Discord, Lark, GitHub, and GitLab, has been launched. It supports ACP-compatible agents such as Claude Code, Codex, Grok Build, DeepSeek, and Pi, allowing them to work together in shared conversations with role-based permissions and memory. The platform is available as a cloud service or self-hosted under the Apache-2.0 license.", "body_md": "**The open-source, multi-agent alternative to Claude Tag.**\n\n@ any agent. Wherever work happens, your agents work alongside\n\nyour team and each other, learning as they go.\n\n** Website** ·\n\n**·**\n\n[Cloud](https://app.agentconnect.md)**·**\n\n[Documentation](https://docs.agentconnect.md)\n\n[Blog](https://agentconnect.md/blog)\n[Why AgentConnect?](#why-agentconnect) ·\n[Get started](#get-started) ·\n[Community](#community) ·\n[Contributing](#contributing) ·\n[Architecture](#architecture)\n\nAgentConnect is an open-source platform where teams and multiple AI agents work together across Slack, Telegram, Discord, Lark, GitHub, and GitLab. Bring Claude Code, Codex, Grok Build, DeepSeek, Pi, or any ACP-compatible agent into the conversations and workflows your team already has open.\n\nGive each agent a role, then let people and agents collaborate in shared conversations. Agents can call one another and remember what they learn, and work can begin from a message, issue, pull request, webhook, or schedule.\n\n▶\n\n**Watch the two-minute introduction**\n\nAI agents are getting better at doing work. The harder problem is making multiple agents work well with a team—and with each other.\n\nMost agents still live like personal tools, in one person's terminal. Teammates can't see what an agent is doing, can't take over a session, can't review its output, and the context it builds stays on one laptop. So every team writes the same glue: message channels, cron jobs, credential handling, context stitching.\n\nAgentConnect turns that glue into a platform.\n\nTeams use it to:\n\n**Triage issues together.** People and agents investigate in one shared thread, bring in the right specialists, and keep the fix and verification visible from start to finish.**Support across trusted workspaces.** Start a support conversation in Telegram, involve engineering from a trusted Slack workspace, and return the resolution where the conversation began.**Run recurring operations.** Start work from a schedule or webhook, bring exceptions into a shared conversation, and keep the human decision visible.**Keep private forks current.** Subscribe to upstream changes through GitHub, a GitHub subscription in Slack, webhooks, or schedules. Let agents assess the impact, prepare and test relevant updates, and bring them to the team for review.**Run customized code review.** Run a general reviewer on every pull request, then bring in architecture or security reviewers only when a change needs them. Each reviewer can use its own model, instructions, repository access, tools, and sandbox policy.\n\n**Work as one team, on any runtime.** Create agents with different roles and let them call on one another, while people follow along in the conversations where the work happens. Claude Code, Codex, Grok Build, DeepSeek, Pi, and any other ACP-compatible runtime run side by side, and changing one does not rebuild the workflow around it.**Keep work where it happens.** Link agents to bots in Slack, Telegram, Discord, and Lark, or to repositories and workflows on GitHub and GitLab.**Choose the right agent for every job.** Configure each agent's runtime, model, workspace, tools, and machine independently.**Carry context forward.** Give each agent its own memory and skills, and publish reviewed[Knowledge](https://docs.agentconnect.md/docs/knowledge)that every agent can find on demand.**Set clear boundaries.** Decide who can see each agent and session, which repositories and tools it may use, and which other agents it may call.**Stay in control.** Self-host the Apache-2.0 stack and keep agent execution and workspaces in the environment you operate.\n\nOne place to operate your whole fleet—agents, sessions, schedules, tools, knowledge, and daemons:\n\n**Don't just take our word for it. Ask your favorite AI about AgentConnect:**\n\nStart the Web console, Control Plane, Relay, and PostgreSQL with Docker Compose:\n\n```\ngit clone https://github.com/agentconnect-md/agentconnect.git\ncd agentconnect\ndocker compose up -d --pull always\n```\n\nOpen `http://localhost:3000`\n\n, add a daemon from the console, run its generated\ncommand, and create your first agent. The default stack listens only on\n`127.0.0.1`\n\nand uses local no-auth mode for evaluation.\n\nThe official Helm chart is published on every release under the same version number as the release itself:\n\n```\noci://ghcr.io/agentconnect-md/charts/agentconnect\n```\n\nA cluster install is three steps — create the namespace and its secrets, write\nthe values file for your deployment, then install — so follow the\n[Kubernetes guide](https://docs.agentconnect.md/docs/kubernetes-deployment)\nrather than a one-liner. The chart source lives in\n[ charts/agentconnect](/agentconnect-md/agentconnect/blob/main/charts/agentconnect).\n\n**Setup Server** runs in the base stack at `http://localhost:8091`\n\n, always on\nloopback. It configures browser authentication through Logto, the deployment's\nGitHub, Slack, Google, and Lark / Feishu apps, the sign-in methods you show, and\npreset-agent behavior. Deployment topology — images, ports, public URLs,\ndatabase and bootstrap secrets — stays in `compose.env`\n\nor your Helm values.\n\nPrefer to be walked through it? This repository ships a **setup skill** at\n`.claude/skills/agentconnect-setup`\n\n, also exposed at `.agents/skills`\n\n, so Claude\nCode, Codex, and other agent harnesses pick it up. Open your coding agent in the\ncheckout and ask it to set up AgentConnect: it runs the guide as an interactive\ntutorial and verifies each checkpoint before continuing, and never asks you to\npaste secrets into chat.\n\nFor authentication, public URLs, Linux sandbox requirements, provider apps,\nimage pinning, secrets, and optional Mem0 configuration, follow the\n[AgentConnect OSS guide](https://docs.agentconnect.md/docs/oss-get-started).\n\nIf AgentConnect looks useful, here's how you can support the project:\n\n- ⭐\nto help more teams discover the project.[Star this repo](https://github.com/agentconnect-md/agentconnect) - 💬\n[Join the Slack community](https://slack.agentconnect.md)for questions, ideas, and product discussions. - 📣 Share AgentConnect on\n[X](https://x.com/intent/post?text=AgentConnect%3A%20the%20open-source%2C%20multi-agent%20alternative%20to%20Claude%20Tag.%20%40%20any%20agent%2C%20wherever%20work%20happens.&url=https%3A%2F%2Fgithub.com%2Fagentconnect-md%2Fagentconnect),[LinkedIn](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fgithub.com%2Fagentconnect-md%2Fagentconnect),[Reddit](https://www.reddit.com/submit?url=https%3A%2F%2Fgithub.com%2Fagentconnect-md%2Fagentconnect&title=AgentConnect%3A%20the%20open-source%2C%20multi-agent%20alternative%20to%20Claude%20Tag), or[LINUX DO](https://linux.do). - ✍️ Write a review or tutorial on\n[DEV](https://dev.to/agentconnect),[Medium](https://medium.com/@agentconnect), or your blog. - 🙋\n[Open an issue](https://github.com/agentconnect-md/agentconnect/issues)to report bugs or suggest features. - 📺 Follow updates on\n[X](https://x.com/getAgentConnect)and[YouTube](https://www.youtube.com/@agentconnect-md).\n\nContributions are welcome. Development needs Node >= 24.12.0 and pnpm 11;\n`pnpm install && pnpm dev`\n\nruns every package in watch mode. The\n[contributing guide](/agentconnect-md/agentconnect/blob/main/CONTRIBUTING.md) covers the full setup, the pull request\nconventions, the monorepo layout, and the deeper design documentation.\n\nThanks to everyone who has already contributed:\n\n| Component | Responsibility |\n|---|---|\nDaemon |\nRuns placed agents over daemon-owned ACP, owns workspaces and session state, maintains direct platform connections and schedules, and sends model-provider traffic directly |\nRelay (optional) |\nAccepts callback-based ingress and webchat, proxies centrally managed MCP and OpenConnector access, and forwards message ingress directly to the owning daemon without durable storage |\nControl Plane + Web UI |\nManages authentication, configuration, placement, permissions, metadata, and observability; stores explicitly approved organization knowledge/skill revisions and otherwise proxies bounded daemon reads on demand |\n\nLive platform messages and ACP update streams stay on the daemon/relay data plane. Apart from explicitly approved organization knowledge and bounded skill bundles, the Control Plane stores coordination metadata—not message bodies, attachment bytes, pending Dream proposals, or ACP session streams. If it is temporarily unavailable, established sessions and daemon-local schedules continue; new assignments and configuration changes resume after reconnection.\n\nSee the\n[architecture design](/agentconnect-md/agentconnect/blob/main/docs/designs/architecture.md) for the\ncomplete message paths, trust boundaries, and failure model.\n\nAgentConnect is available under the [Apache License 2.0](/agentconnect-md/agentconnect/blob/main/LICENSE).", "url": "https://wpnews.pro/news/show-hn-agentconnect-shared-agents-with-separate-permissions", "canonical_source": "https://github.com/agentconnect-md/agentconnect", "published_at": "2026-08-26 13:58:08+00:00", "updated_at": "2026-08-26 14:16:11.290211+00:00", "lang": "en", "topics": ["ai-agents", "ai-tools", "developer-tools"], "entities": ["AgentConnect", "Claude Code", "Codex", "Grok Build", "DeepSeek", "Pi", "Slack", "Telegram"], "alternates": {"html": "https://wpnews.pro/news/show-hn-agentconnect-shared-agents-with-separate-permissions", "markdown": "https://wpnews.pro/news/show-hn-agentconnect-shared-agents-with-separate-permissions.md", "text": "https://wpnews.pro/news/show-hn-agentconnect-shared-agents-with-separate-permissions.txt", "jsonld": "https://wpnews.pro/news/show-hn-agentconnect-shared-agents-with-separate-permissions.jsonld"}}