Agent Tincan: Open-Source Tool Lets Your AI Agents Talk to Each Other Over Tailscale
Scobleizer · x · 2026-09-26
A developer released Agent Tincan, an open-source project addressing a common pain point: Grok, ChatGPT, Claude, Codex and other agents each work in silos, leaving the user as the copy-paste middleman.
- Each agent has a superpower: always-on cloud Grok Bot, phone-calling Muse, Instinct in iMessage, Codex/Claude Code with your code
- Agents connect over a private Tailscale network — no open ports, nothing leaves it
- Example flows: have Grok Bot make Muse call a restaurant, pull a ChatGPT-generated image from any agent, retrieve details from an old Claude chat, talk to Codex on your dev machine
- Agent-first setup: paste one message into your always-on agent and it configures the rest
More from coding & agent
- wc3env: Warcraft 3 Frozen Throne turned into an open-source RL environment — daveholtz · 2026-09-26
- Cua ships stable Omarchy driver with native synthetic cursors for multi-cursor computer use — EdwardSun0909 · 2026-09-26
- Codex takes 15 minutes to fix a bug, then 10 more just to explain what it did — burkov · 2026-09-26
- Dev enforces screenshot proof on every frontend PR via an agent skill — SeanOliver · 2026-09-26
- Gate the action, not the confidence score: why agent risk control should follow capabilities — arslannasir128 · 2026-09-26
- Researcher's agent workflow constraints: no terminal, seamless across three machines and iPhone — sethlazar · 2026-09-26