ArchAstro

ArchAstro

Developer Docs

Build agents your team can ship with confidence

Agents that keep their identity, tools, knowledge, and conversation history. Deploy from the CLI or your coding agent. Collaborate across companies when the work crosses boundaries.

Getting startedGitHub ReleasesPortal

Build

Deploy your first agent

Write an agent.yaml, deploy it, and test it in a real conversation from the CLI.

Operate

Inspect, test, and iterate

Use the CLI to inspect runs, debug tools, test in sandboxes, and connect knowledge — all from the terminal.

Collaborate

Connect across companies

Keep each company private by default, then create one shared team and thread when collaboration needs to cross the boundary.

Using Claude Code, Codex, or Gemini CLI?

Paste this into your coding agent to get the repo linked, the CLI installed, and one agent deployed quickly.

ClaudeCodexGemini
Set up ArchAstro in this repo so we can deploy an agent and test it.

1) Read:
   https://docs.archastro.ai/llms-full.txt

2) Ask me for any missing ArchAstro credentials or environment variables.

3) Install the ArchAstro CLI and run: archastro auth login && archastro init

4) Install a starter agent from the catalog: archastro list agentsamples, then archastro install agentsample <slug>. (If I ask for something specific instead, write an agent.yaml — kind: AgentTemplate, clear identity, participate preset routine, search + knowledge_search builtin tools, memory/long-term installation — and deploy with archastro deploy agent agent.yaml.)

5) Test it: create an agent session and exec a test prompt, or create a thread + user + message.

6) If I want to embed the agent locally, remind me: Claude Code → /agents:embed, Codex → $archagents:embed.

7) When complete, summarize what was created and how to test it again.

Need the fuller version? Open the coding-agent guide.